Publications

A-UVI: GNSS-Assisted EO-based UV Index Estimation Method for Individual-level Precise UV Exposure Assessment

Yuuki Nishiyama, Subaru Atsumi, Kota Tsubouchi, Kaoru Sezaki

Proc. ACM Interact. Mob. Wearable Ubiquitous Technol.

Excessive or insufficient exposure to ultraviolet (UV) light can have adverse effects on health, including the development of skin cancer, cataracts, and osteoporosis. An Earth observation (EO)-based UV index can estimate area-level UV indexes without effort in open-sky environments but can not provide sufficient accuracy for shaded environments. In contrast, conventional methods for monitoring individual-level, i.e., personal, UV exposure, such as mobile and wearable UV sensors, face limitations in terms of measurement and usability, presenting challenges for practical long-term usage. To address these issues, we introduce A-UVI, a method that enhances the accuracy of the EO-based UV index by leveraging raw signals from global navigation satellite systems (GNSS). By integrating this EO-based UV index and an attenuation ratio estimated from raw GNSS signals, our method especially improves estimation accuracy in shady environments affected by obstructions. We evaluated our method on data collected by different GNSS receivers in different mobility scenarios encompassing a diverse range of contexts and observation areas over the course of three days. Our evaluation showed that A-UVI estimates the UV index with a precision exceeding existing methods by at least 44.25%, achieving 5.53 times higher estimation accuracy in forest environments. We also confirmed that A-UVI is compatible with GNSS receivers in consumer-grade smartphones and has an average accuracy that is 23% better than the baseline EO-based method. Our findings demonstrate that utilizing raw GNSS signals enables accurate estimation of the UV index in various conditions, including in shaded areas, without the need for particular measurement actions or devices. This marks a significant advancement in enabling passive individual-level UV exposure monitoring and adaptive UV exposure management beyond simple exposure tracking.

HeadMon+: Domain Adaptive Head Dynamic-based Riding Maneuver Prediction

Zengyi Han, En Wang, Mohan Yu, Jie Wang, Yuuki Nishiyama, Kaoru Sezaki

IEEE Transactions on Mobile Computing

Micro-mobility has become a vital means of transportation in recent years, however, it has also resulted in a rise in traffic incidents. Timely tracking and predicting riders' maneuvers hold the potential to ensure active protection and allow for sufficient time to avert accidents by issuing timely warnings and interventions. We contend that the rider's head dynamics can provide valuable information regarding their subsequent maneuvers. Riders' traveling habits, however diverse, not to mention the rapidly varying riding environment. The above factors contribute to significant disruptions in the data source, and various micro-mobility forms further exacerbate the issue. We accordingly present HeadMon+, which predicts the rider's subsequent maneuver by examining their head dynamics, and it can effectively adapt to various riding conditions and individuals. The system incorporates a deep learning framework with an advanced domain adversarial network. By single-time pre-training, HeadMon+ is capable of adapting to new data domains, including human subjects, and riding conditions for robust maneuver prediction. Based on our evaluation, we have found that the maneuver prediction of HeadMon+ has an overall precision of 94% with a prediction time gap of 4 seconds. HeadMon+'s low cost and rapid response capability make it easily deployed and then contribute to enhancing safe riding.

Image Generative Semantic Communication with Multi-Modal Similarity Estimation for Resource-Limited Networks

Eri Hosonuma, Taku Yamazaki, Takumi Miyoshi, Akihito Taya, Yuuki Nishiyama, Kaoru Sezaki

IEICE Transactions on Communications

Adaptive Detection of Spread Spectrum Signals With General Array Configuration

Yutong Feng, Akihito Taya, Yuuki Nishiyama, Jun Liu, Kaoru Sezaki

IEEE Transactions on Signal Processing

Multi-label Classification Model for Infant Activity Recognition Using Single Inertial Sensor

Ayaka Onodera, Riku Ishioka, Yuuki Nishiyama, Kaoru Sezaki

IEEE Pervasive Computing

Recording and sharing childcare information is crucial for accurately assessing a child's health status and taking appropriate action in case of illness or other emergencies. Although numerous applications and systems have been proposed to assist in recording and sharing these records, the process is still performed manually, presenting a significant burden for parents. Therefore, automatic recording of infants' daily activities is required. In this study, we implement a machine learning model to recognize multi-labeled infant activities using a chest-mounted low-sampling rate accelerometer. We collected accelerometer data from twenty-four infants between 6 and 24 months as a dataset. Based on the data, we extracted 25 time- and frequency-domain features calculated from the single accelerometer and user features to recognize the fourteen daily activities. The performance evaluation considering multi-label classification showed that our proposed model reaches over 88% in the F1 score in the best case.

AMoND: Area-Controlled Mobile Ad-hoc Networking with Digital Twin

Shota Ono, Taku Yamazaki, Takumi Miyoshi, Akihito Taya, Yuuki Nishiyama, Kaoru Sezaki

IEEE Access

Future smart cities are expected to provide intelligent services such as predictions, detections, and automation through digital twins. However, the creation of digital twins requires the processing of an enormous amount of data, thereby leading to an increase in mobile network traffic. This traffic is produced by applications in user devices and city services, which engage in local consumption at the city scale through sensor and camera devices using mobile networks. Such increased traffic can compromise the communication speed and stability. To alleviate this burden, traffic offloading becomes a crucial consideration in the beyond-5G era. This paper presents a scheme known as Area-Controlled Mobile Ad-Hoc Networking (AMoND). AMoND uses a hierarchical structure of a location layer and an ad-hoc layer to construct area-controlled mobile ad-hoc networks (MANETs) for mutual support of the digital twin and MANETs. AMoND effectively suppresses mobile network traffic by harnessing the digital twin to assist the MANETs during data collection for the digital twin construction. Importantly, the digital twin used in AMoND focuses on the management of node location information and does not need to reproduce the real space on a computer fully. AMoND is not dependent on a specific MANET protocol and can be used as an add-on. AMoND exhibits the ability to reduce traffic volumes by up to approximately 65%, while maintaining arrival rates that are comparable to existing MANET protocols under certain conditions.

Compressive Detection of Stochastic Sparse Signals With Unknown Sparsity Degree

Yutong Feng, Akihito Taya, Yuuki Nishiyama, Kaoru Sezaki, Jun Liu

IEEE Signal Processing Letters

In this letter, we investigate the problem of detecting compressed stochastic sparse signals with unknown sparsity degree under Bernoulli–Gaussian model. In addition to the generalized likelihood ratio test (GLRT) proposed in (Hariri and Babaie–Zadeh et al., 2017), the corresponding Rao test and Wald test are derived in this letter. By observing that obtaining their analytical performance is challenging, we further propose a new probability constraint estimator (PCE) of the unknown sparsity degree. Interestingly, by adopting the PCE, the GLRT, Rao and Wald tests are shown to be statistically equivalent and reduce to a new detector (i.e., the detector with PCE) with a simple structure. The analytical performance of the detector with PCE is thus derived, which is verified by Monte Carlo simulations. Finally, numerical experiments illustrate that the proposed Rao test and the detector with PCE outperform the original GLRT.

Assessing environmental benefits from shared micromobility systems using machine learning algorithms and Monte Carlo simulation

Helinyi Peng, Yuuki Nishiyama, Kaoru Sezaki

Sustainable Cities and Society

Shared micromobility systems (SMSs) are paving the way for new, more convenient travel options while also lowering transportation-related greenhouse gas (GHG) emissions. However, few studies have used real-world trip data to estimate SMSs' environmental benefits, especially for dockless scooter-sharing services. To this end, we proposed a system to estimate the GHG emission reduction effected by SMSs. First, several machine learning (ML) algorithms were utilized to identify citizens' travel mode choice preferences, and then the mode substituted by each shared micromobility trip was estimated. We compared the ML algorithms' estimation results and selected those from the random forest, lightGBM, and XGBoost model for further estimating GHG reductions. Second, the Monte Carlo simulations were used to simulate the substituted mode at the trip level to improve the reliability of the final GHG reduction estimation. Finally, the environmental benefits were calculated based on the trip distances and the travel modes that were substituted. Instead of estimating a specific number, we obtained a probabilistic outcome for the environmental benefits while considering the level of uncertainty. Our results suggest that SMSs have positive environmental impacts and have the potential to facilitate the decarbonization of urban transport. According to these findings, implications and suggestions on extending SMSs' environmental benefits are proposed.

SFC GO: 学生同士の繋がりを支援するオンライン体育授業サポートシステム

佐々木航, 柿野優衣, 中縁嗣, 野田悠加, 羽柴彩月, 山田佑亮, 西山勇毅, 大越匡, 中澤仁, 森将輝, 水鳥寿思, 塩田琴美, 永野智久, 東海林祐子, 加藤貴昭

情報処理学会論文誌デジタルプラクティス（TDP）

COVID-19感染拡大の影響を受け，慶應義塾大学では2020年春学期のすべての授業が，体育も含めてオンライン開催となった．特に大学新入生が全員履修する「体育1」は身体運動体験を通じたクラスメート同士のコミュニケーションの場であり，オンライン授業においてもその機会の損失を防ぐことが求められた．そこで我々は情報系教員・学生と体育教員の知見を融合させた，オンライン体育授業サポートシステム「SFC GO」を1ヶ月で構築し運用した．SFC GOではスマートフォン内蔵センサを用いた身体運動の記録や振り返りが可能であり，出題される課題に則した運動記録をタイムラインへ投稿できる．また，クラスメートの投稿を閲覧することや投稿へコメントすることなどのソーシャルネットワーク機能を有する．そして，バックグラウンドで定常的にセンサデータを収集することによって，家事や散歩などの授業時間以外の日常的な運動の記録や振り返りができる．体育1を履修した学生を対象に学生自身のスマートフォンに本アプリケーションをインストールし，2020年5月から7月のオンライン体育授業で使用した．本稿では，実施期間で収集された運動記録のデータ，テクニカルサポート対応事例やスマートフォンを活用した本実施の経験などから導かれる知見について考察を行う．

感染症流行時におけるスマートフォンを用いた大学生の身体活動量分析

西山勇毅, 柿野優衣, 中縁嗣, 野田悠加, 羽柴彩月, 山田佑亮, 佐々木航, 大越匡, 中澤仁, 森将輝, 水鳥寿思, 塩田琴美, 永野智久, 東海林祐子, 加藤貴昭

情報処理学会論文誌 [特集：ユビキタスコンピューティングシステム（X）]

新型コロナウイルス感染症（COVID-19）の世界的な感染拡大にともない，多くの大学ではキャンパス内での感染予防のために，キャンパスの封鎖とインターネット越しに授業を配信するオンライン授業が導入され，学生たちは自宅から授業に参加している．このような在宅中心の新しい生活様式は，感染予防効果が見込める一方で，運動不足による二次的な健康被害が懸念される．新しい生活様式における大学生の身体活動の実態，特に学生の属性や時間帯ごとの身体活動量とその内容を明らかにすることは，二次的な健康被害を予防するうえで必要不可欠である．そこで本研究では，日常生活中の身体活動データ（歩数と6種類の行動種別）を大学生が所有するスマートフォンを用いて自動収集し，大学生の身体活動量を明らかにする．身体活動データは，必修の体育授業を履修する大学1年生305名から10週間収集した．その結果，通学（7時から10時）や教室での授業，課外活動（11時から24時）の時間帯における歩数の減少と静止時間の長時間化が明らかになった．本結果は，新しい生活様式における大学生活が平日の身体活動量の低下を招く可能性を示唆する． With the spreading of the new coronavirus infection (COVID-19) worldwide, several universities have closed their campuses to prevent the spread of infection. Consequently, university classes are being held over the Internet, and students attend these classes from their homes. While the COVID-19 pandemic is expected to be prolonged, the online-centric lifestyle has raised concerns about secondary health issues caused by reduced physical activity (PA). However, the actual status of PA among college students has not yet been examined in Japan. Hence, in this study, we collected daily PA data (including the data corresponding to the number of steps taken and the data associated with six types of activities) by employing off-the-shelf smartphones and thereby analyzing the PA changes of college students. The PA data were collected over a period of ten weeks from 305 first-year college students who were attending a mandatory class of physical education at the university. The obtained results indicate that the decrease in commuting time (7 AM to 10 AM), classroom time, and extracurricular activity time (11 AM to 12 AM) has led to a decrease in PA on weekdays owing to reduced unplanned exercise opportunities. The results suggest that college life in an online-centric lifestyle may lead to a decrease in PA on weekdays.

MOCHA: Bluetoothビーコンを用いた学内位置情報サービスの開発・運用

西山勇毅, 川原圭博, 瀬崎薫

画像電子学会誌

新型コロナウイルス感染症（COVID-19）の世界的な感染拡大により，感染症予防策の一つとして多くの大学では，オンライン授業や対面とオンライン授業を併用するハイブリッド授業を用いて教育・研究機会を提供してきた．2021年に入ってからは，多くの大学では，完全オンライン授業からハイブリッド・対面授業に移行しつつある．しかし，COVID-19 は常に変異しており，感染症の再流行も懸念されている．本格的な対面授業や研究活動の再開に向けて，学内管理施設における安心・安全の確保のためには，施設内における人々の滞在状況や人流・接触状況などを把握できることが望ましい．また，これらの情報は，アフターコロナにおいても，部屋の予約や道案内，場所に応じたリマインド機能などの DX 基盤としても活用できる．一方で，位置情報利用におけるプライバシーの確保やインセンティブ設計，開発・運用体制など実現には，技術以外にも多くの課題が存在する．

Investigating the occurrence of selfie-based emotional contagion over social network

Wataru Sasaki, Yuuki Nishiyama, Tadashi Okoshi, Jin Nakazawa

Social Network Analysis and Mining

Happiness is obviously one of the most fundamental essence that affects many aspects of our lives. Past research found that happiness of one person affects that of other people. What occurs under this propagation of emotion is called “emotional contagion,” a phenomenon wherein through perception, people experience the same emotion expressed by someone when communicating with them. Although online communication is increasing due to growth of mobile computing, emotional contagion on online communication is not well studied yet. Particularly, it is not yet clear if emotional contagion among people occurs through selfie photographs posted on the social network media. We implemented “SmileWave,” the social networking system for investigating selfie-based emotional contagion. The key feature of SmileWave is detecting “smile degree” in user’s posting selfies and in reactive facial expressions when the user is viewing the posted photographs from others. Our in-the-wild user studies with 38 participants for 2 weeks revealed the occurrence of selfie-based emotional contagion over the social network, based on the results that the users’ smile degree improved (15% on average) when the user looked at posted selfie photographs.

Mobile phone sensor-based detection of subjective cannabis intoxication in young adults: A feasibility study in real-world settings

Sang Won Bae, Tammy Chung, Rahul Islam, Brian Suffoletto, Jiameng Du, Serim Jang, Yuuki Nishiyama, Raghu Mulukutla, Anind Dey

Drug and Alcohol Dependence

Background Given possible impairment in psychomotor functioning related to acute cannabis intoxication, we explored whether smartphone-based sensors (e.g., accelerometer) can detect self-reported episodes of acute cannabis intoxication (subjective “high” state) in the natural environment. Methods Young adults (ages 18–25) in Pittsburgh, PA, who reported cannabis use at least twice per week, completed up to 30 days of daily data collection: phone surveys (3 times/day), self-initiated reports of cannabis use (start/stop time, subjective cannabis intoxication rating: 0–10

Smartphone-Based Monitoring of Parkinson Disease: Quasi-Experimental Study to Quantify Hand Tremor Severity and Medication Effectiveness

Elina Kuosmanen, Florian Wolling, Julio Vega, Valerii Kan, Yuuki Nishiyama, Simon Harper, Kristof Van Laerhoven, Simo Hosio, Denzil Ferreira

JMIR Mhealth Uhealth

Background: Hand tremor typically has a negative impact on a person's ability to complete many common daily activities. Previous research has investigated how to quantify hand tremor with smartphones and wearable sensors, mainly under controlled data collection conditions. Solutions for daily real-life settings remain largely underexplored. Objective: Our objective was to monitor and assess hand tremor severity in patients with Parkinson disease (PD), and to better understand the effects of PD medications in a naturalistic environment. Methods: Using the Welch method, we generated periodograms of accelerometer data and computed signal features to compare patients with varying degrees of PD symptoms. Results: We introduced and empirically evaluated the tremor intensity parameter (TIP), an accelerometer-based metric to quantify hand tremor severity in PD using smartphones. There was a statistically significant correlation between the TIP and self-assessed Unified Parkinson Disease Rating Scale (UPDRS) II tremor scores (Kendall rank correlation test: z=30.521, P<.001, $tau$=0.5367379; n=11). An analysis of the ``before'' and ``after'' medication intake conditions identified a significant difference in accelerometer signal characteristics among participants with different levels of rigidity and bradykinesia (Wilcoxon rank sum test, P<.05). Conclusions: Our work demonstrates the potential use of smartphone inertial sensors as a systematic symptom severity assessment mechanism to monitor PD symptoms and to assess medication effectiveness remotely. Our smartphone-based monitoring app may also be relevant for other conditions where hand tremor is a prevalent symptom.

Mobile Assessment of Acute Effects of Marijuana on Cognitive Functioning in Young Adults: Observational Study

Tammy Chung, Sang Won Bae, Eun-Young Mun, Brian Suffoletto, Yuuki Nishiyama, Serim Jang, Anind K Dey

JMIR Mhealth Uhealth

Background: Mobile assessment of the effects of acute marijuana on cognitive functioning in the natural environment would provide an ecologically valid measure of the impacts of marijuana use on daily functioning. Objective: This study aimed to examine the association of reported acute subjective marijuana high (rated 0-10) with performance on 3 mobile cognitive tasks measuring visuospatial working memory (Flowers task), attentional bias to marijuana-related cues (marijuana Stroop), and information processing and psychomotor speed (digit symbol substitution task [DSST]). The effect of distraction as a moderator of the association between the rating of subjective marijuana high and task performance (ie, reaction time and number of correct responses) was explored. Methods: Young adults (aged 18-25 years; 37/60, 62% female) who reported marijuana use at least twice per week were recruited through advertisements and a participant registry in Pittsburgh, Pennsylvania. Phone surveys and mobile cognitive tasks were delivered 3 times per day and were self-initiated when starting marijuana use. Completion of phone surveys triggered the delivery of cognitive tasks. Participants completed up to 30 days of daily data collection. Multilevel models examined associations between ratings of subjective marijuana high (rated 0-10) and performance on each cognitive task (reaction time and number of correct responses) and tested the number of distractions (rated 0-4) during the mobile task session as a moderator of the association between ratings of subjective marijuana high and task performance. Results: Participants provided 2703 data points, representing 451 reports (451/2703, 16.7%) of marijuana use. Consistent with slight impairing effects of acute marijuana use, an increase in the average rating of subjective marijuana high was associated with slower average reaction time on all 3 tasks---Flowers (B=2.29; SE 0.86; P=.008), marijuana Stroop (B=2.74; SE 1.09; P=.01), and DSST (B=3.08; SE 1.41; P=.03)---and with fewer correct responses for Flowers (B=−0.03; SE 0.01; P=.01) and DSST (B=−0.18; SE 0.07; P=.01), but not marijuana Stroop (P=.45). Results for distraction as a moderator were statistically significant only for certain cognitive tasks and outcomes. Specifically, as hypothesized, a person's average number of reported distractions moderated the association of the average rating of subjective marijuana high (over and above a session's rating) with the reaction time for marijuana Stroop (B=−52.93; SE 19.38; P=.006) and DSST (B=−109.72; SE 42.50; P=.01) and the number of correct responses for marijuana Stroop (B=−0.22; SE 0.10; P=.02) and DSST (B=4.62; SE 1.81; P=.01). Conclusions: Young adults' performance on mobile cognitive tasks in the natural environment was associated with ratings of acute subjective marijuana high, consistent with slight decreases in cognitive functioning. Monitoring cognitive functioning in real time in the natural environment holds promise for providing immediate feedback to guide personal decision making.

HealthyStadium: 他者評価とゲーミフィケーションを用いた食習慣改善ソーシャルメディア

栄元優作, 西山勇毅, 大越匡, 中澤仁

情報処理学会論文誌

生活習慣病は，生活習慣が発症原因に深く関与する疾患の総称であり，癌などの三大死因の発症リスクを高める危険性がある．特に食事は，人間活動に必須であり日常的に行われるが，高カロリー・栄養バランスの悪い食事は生活習慣病につながる可能性が高い．健康な食生活を目指すために，主にモバイルデバイスに食事内容を記録する手法が利用されるが，栄養素の計算・記録負荷が大きく，また利用者がつねに能動的に記録を行う必要があり，継続利用には利用者個人の動機に依存する傾向がある．本研究では，これまでの食事内容の記録手法に「グループでのゲーム要素」および「食事内容に対する他者の評価」を加えることで，グループ全体での低負荷での食事の行動変容を促進する新たな記録手法，HealthyStadiumを提案する．HealthyStadiumはグループ内において食事画像の共有および，その健康度合いで対戦するソーシャルメディアである．健康度合いの評価は，1人1票の投票形式で行われ投票数の多い方が勝利となる．本アプリケーションを用いて，大学生10人を対象に28日間にわたる評価実験を実施した．既存の記録アプリケーションと比較して，HealthyStadiumでは食事の記録回数が20%上昇し，記録負担が低減した．さらに，行動変容ステージモデルの評価指数の1つである自己効力感を増加させる効果を確認できた． Lifestyle-related diseases are a general term for diseases in which lifestyle is deeply involved in the pathogenesis, and there is a risk of raising the risk of developing three major causes of death such as cancer. Diet, in particular, is essential for human activities and is carried out on a daily basis, but diets with high calorie and poor nutritional balance are likely to lead to lifestyle diseases. To aim for healthy eating habits, a method of recording dietary contents in mobile devices is mainly used, but calculation of nutrients and recording load are large and it is required to always positively record, so there is a tendency to depend on the motivation of individual users for continuous use. In this research, we propose “HealthyStadium”, a new recording method to promote eating behavior change in the whole group by adding “game element in group” and “evaluation of others to meal contents” to the existing recording method. HealthyStadium is a social media which shares meal photo in the group and makes users fight with its health degree. Evaluation of the degree of health is done by voting one by one person, and users with many votes win. Using this application, we conducted an evaluation experiment for 10 university students over 28 days. Compared with the existing recording application, the number of meals recordings increased by 20% in HealthyStadium, and the burden of the recording was alleviated. Furthermore, the self-efficacy which is one of the evaluation indexes of the behavior change stage model was also improving.

A Platform for Mutual Watch-Over among the Elderly Using PAN and Gamification

Jin Nakazawa, Wataru Sasaki, Mikio Obuchi, Kazuki Egashira, Yuuki Nishiyama, Tadashi Okoshi, Takuro Yonezawa, Hideyuki Tokuda

The IEICE transactions on information and systems

This paper proposes a platform, called Kugenuma Happy Board, that strengthens communication among elderly in local community for mutual watch-over among them. The platform motivates users to join the mutual watch-over leveraging health-related data acquired from a personal area network (PAN), sharing selfy pictures of smiley face among community, and gamification techniques including missions, scoring, and rankings. We have conducted an in-the-wild experiment for 10 months in Fujisawa Kanagawa, Japan. The experiment showed that the platform accelerates information sharing thus mutual watch-over among users. Particularly, we have observed that elderly users perceived other users activity from shared information, such as step counts and selfy pictures, thus strengthened human relationship in the local community.

MyFactor：ユーザの内面状態に寄与する因子情報の個人特性に適応的な分析

小渕幹夫, 西山勇毅, 大越匡, 米澤拓郎, 中澤仁

情報処理学会論文誌

近年，様々なセンサを搭載したスマートフォンの発達と普及により，第三者による客観視が困難である「忙しい」や「疲れている」といった人間の内面状態の推定を行う研究が行われている．人間の内面状態を機械に認識させ，内面状態に寄与している因子に関する情報を人々に還元できれば，人間が自身の感情や気分を制御したり，特定の内面状態の獲得や回避といった行動変容へと応用可能である．しかし，人間の内面状態の認識・分析・還元においては，各個人の多様性に適応した動作の実現が課題である．本研究ではスマートフォンのセンサデータを蓄積し，正解データとなる内面状態の注釈をユーザが記録することで，個人の内面状態に寄与している因子を還元するために，因子情報を解析するシステム「MyFactor」を構築した．大学生13名による111日にわたる評価実験を行った結果，被験者全員のデータを用いて学習を行う「統合モデル」と比較して，各ユーザのデータのみを用いて個別に学習した「個人モデル」においては内面状態の推定精度が21%上昇した．また，利用可能なすべてのセンサからのデータを用いて学習した場合よりも，単一センサからのデータのみを用いて学習を行い，最も高精度で分類を行ったモデルを選択した方が高い推定精度を記録し，ユーザに自身の内面状態に関して深い理解を与える可能性を示した． Recently, due to the advance and spread of smartphones equipped with various sensors, human beings' inner states such as "busyness" or "tiredness" can be detected more accurately. If people can recognize the factors contributing to their inner states, they will be able to regulate their own emotions and moods by paying attention to those factors. However, the adaptive system for each individual's diversity that recognize, analyze, and provide the factors contributing to our inner state is not realized. In this research, we propose a system, MyFactor, that accumulates smartphone sensor data and user annotations about inner states to extract and provide information about the important factors contributing to a user's inner states. In a 111-day experiment with 13 university students, researchers estimated that an "individual model" developed only with the data of each user had a higher predictive accuracy for human inner states than the generic model developed using data from all subjects. In addition, the model with the highest predictive accuracy, trained by a single sensor, performed better and showed the possibility of giving a deeper understanding of human inner states than the model trained using data from all available sensors.

Toward Health Exercise Behavior Change for Teams Using Lifelog Sharing Models

Yuuki Nishiyama, Tadashi Okoshi, Takuro Yonezawa, Jin Nakazawa, Kazunori Takashio, Hideyuki Tokuda

IEEE Journal of Biomedical and Health Informatics

Recent technological trends in mobile/wearable devices and sensors have been enabling an increasing number of people to collect and store their “lifelog” easily in their daily lives. Beyond exercise behavior change of individual users, our research focus is on the behavior change of teams, based on lifelogging technologies and lifelog sharing. In this paper, we propose and evaluate six different types of lifelog sharing models among team members for their exercise promotion, leveraging the concepts of “competition” and “collaboration.” According to our experimental mobile web application for exercise promotion and an extensive user study conducted with a total of 64 participants over a period of three weeks, the model with a “competition” technique resulted in the most effective performance for competitive teams, such as sports teams.

ライフログデータを用いたチームの行動変容促進

西山勇毅, 大越匡, 米澤拓郎, 中澤仁, 高汐一紀, 徳田英幸

情報処理学会論文誌

近年，携帯端末の普及にともない誰もが日常生活中の活動情報を検知・蓄積可能なライフログ環境が整ってきた．これまでライフログデータは，個人を対象として行動変容の促進に活用されてきたが，今後は研究室やスポーツチーム，企業といった集団を対象とした行動変容の促進が可能になると考えられる．しかし，集団は個人とは異なり内部に様々な人間関係が存在するため，これまでの個人を対象とした行動変容促進手法が集団に対して効果的であるかは明らかではない．本研究では，集団の行動変容を促進するモデルとして，既存手法の「競争」と「協力」の要素を組み合わせた6種類の集団の行動変容促進モデルの提案し，効果の検証を行った．提案モデルに基づいた行動変容促進を行うAaron2を実装し，2つの集団（64名）を対象に3週間の実験を行った．1週間ごとの行動変容について考察した結果，チーム目標と直接的に関係ない活動では行動変容への効果が低く，日頃からチーム単位で競争を行っているチームでは「チーム間での競争要素」用いたモデルが最も行動変容への効果が高くなる可能性が示された．Recent technological trends on mobile/wearable devices and sensors have been enabling increasing number of people to collect and store their "lifelog" easily in their daily lives. Beyond exercise behavior change of individual user, our research focus is on the behavior change of teams, based on life-logging technologies and information sharing. In this paper, we propose and evaluate six different types of information sharing model among team members for their exercise promotion, leveraging concepts of "competition" and "collaboration". According to our experimental mobile web application for exercise promotion and extensive user study among 64 total users for three weeks, the model with "competition" technique resulted the most effective performance for competitive teams such as sport teams. Recent technological trends on mobile/wearable devices and sensors have been enabling increasing number of people to collect and store their "lifelog" easily in their daily lives. Beyond exercise behavior change of individual user, our research focus is on the behavior change of teams, based on life-logging technologies and information sharing. In this paper, we propose and evaluate six different types of information sharing model among team members for their exercise promotion, leveraging concepts of "competition" and "collaboration". According to our experimental mobile web application for exercise promotion and extensive user study among 64 total users for three weeks, the model with "competition" technique resulted the most effective performance for competitive teams such as sport teams.

RideStyle: Riding Style Representation from Head-Body Dynamics via Adversarial Learning

Zengyi Han, Yuuki Nishiyama, Kaoru Sezaki

EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services

Deep Learning-Based Compressed Sensing for Mobile Device-Derived Sensor Data

Liqiang Xu, Yuuki Nishiyama, Kota Tsubouchi, Kaoru Sezaki

Proceedings of the 33rd ACM International Conference on Information and Knowledge Management

As the capabilities of smart sensing and mobile technologies continue to evolve and expand, storing diverse sensor data on smartphones and cloud servers becomes increasingly challenging. Effective data compression is crucial to alleviate these storage pressures. Compressed sensing (CS) offers a promising approach, but traditional CS methods often struggle with the unique characteristics of sensor data-like variability, dynamic changes, and different sampling rates-leading to slow processing and poor reconstruction quality. To address these issues, we developed Mob-ISTA-1DNet, an innovative CS framework that integrates deep learning with the iterative shrinkage-thresholding algorithm (ISTA) to adaptively compress and reconstruct smartphone sensor data. This framework is designed to manage the complexities of smartphone sensor data, ensuring high-quality reconstruction across diverse conditions. We developed a mobile application to collect data from 30 volunteers over one month, including accelerometer, gyroscope, barometer, and other sensor measurements. Comparative analysis reveals that Mob-ISTA-1DNet not only enhances reconstruction accuracy but also significantly reduces processing time, consistently outperforming other methods in various scenarios.

RideGuard: Micro-Mobility Steering Maneuver Prediction with Smartphones

Zengyi Han, Xuefu Dong, Liqiang Xu, Zhen Zhu, En Wang, Yuuki Nishiyama, Kaoru Sezaki

2024 IEEE 44th International Conference on Distributed Computing Systems (ICDCS)

Although micro-mobility has become a popular and indispensable mode of transportation in recent years, it has also introduced a large number of traffic accidents. Timely tracking and predicting the maneuvers hold the potential to prevent accidents through prompt warnings and interventions. However, the open and simple structure of micro-mobility makes it hard to install sophisticated infrastructures for maneuver prediction. In this paper, we argue that the micro-mobility body dynamics provide sufficient information for maneuver prediction. Our preliminary study suggests that micro-mobility body dynamic patterns appear beforehand and exhibit the correlation with steering maneuvers. We accordingly present RideGuard, which leverages a built-in Inertial Measurement Unit on smartphones to achieve the prediction of steering maneuvers. Through a dual- stream CNN deep learning architecture, RideGuard effectively captures complex patterns and feature relationships from the time and frequency domain. Our extensive real-traffic experi- ments involving 20 participants demonstrate the superiority of RideGuard: employing a 3s detection window, RideGuard attains a minimum of 94% precision in maneuver prediction with a 5s prediction time gap. The low-cost and rapid response feature of RideGuard enables feasible deployment and promotes safer riding practices. Additionally, we open-source our well-labeled dataset to facilitate further research.

ReHEarSSE: Recognizing Hidden-in-the-Ear Silently Spelled Expressions

Xuefu Dong, Yifei Chen, Yuuki Nishiyama, Kaoru Sezaki, Yuntao Wang, Kenneth Christofferson, Alex Mariakakis

Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems

Silent speech interaction (SSI) allows users to discreetly input text without using their hands. Existing wearable SSI systems typically require custom devices and are limited to a small lexicon, limiting their utility to a small set of command words. This work proposes ReHEarSSE, an earbud-based ultrasonic SSI system capable of generalizing to words that do not appear in its training dataset, providing support for nearly an entire dictionary’s worth of words. As a user silently spells words, ReHEarSSE uses autoregressive features to identify subtle changes in ear canal shape. ReHEarSSE infers words using a deep learning model trained to optimize connectionist temporal classification (CTC) loss with an intermediate embedding that accounts for different letters and transitions between them. We find that ReHEarSSE recognizes unseen words with an accuracy of pmnice{89.3}{10.9}%.

An Optimization Method for Indoor Trajectory Estimation from Spatially Sparse and Noisy Beacon Data

Junya Maruyama, Yudai Honma, Yuuki Nishiyama, Yoshihiro Kawahara

OHOW 2023 – The 2nd International Symposium on One Health / Infrastructure Management and sustainable built environment

Convergence Visualizer of Decentralized Federated Distillation with Reduced Communication Costs

Akihito Taya, Yuuki Nishiyama, Kaoru Sezaki

Proceedings of IEEE Global Communications (GLOBECOM) 2023

Federated learning (FL) achieves collaborative learning without the need for data sharing, thus preventing privacy leakage. To extend FL into a fully decentralized algorithm, researchers have applied distributed optimization algorithms to FL by considering machine learning (ML) tasks as parameter optimization problems. Conversely, the consensus-based multi-hop federated distillation (CMFD) proposed in the authors' previous work makes neural network (NN) models get close with others in a function space rather than in a parameter space. Hence, this study solves two unresolved challenges of CMFD: (1) communication cost reduction and (2) visualization of model convergence. Based on a proposed dynamic communication cost reduction method (DCCR), the amount of data transferred in a network is reduced; however, with a slight degradation in the prediction accuracy. In addition, a technique for visualizing the distance between the NN models in a function space is also proposed. The technique applies a dimensionality reduction technique by approximating infinite-dimensional functions as numerical vectors to visualize the trajectory of how the models change by the distributed learning algorithm.

HeadSense: Visual Search Monitoring and Distracted Behavior Detection for Bicycle Riders

Zengyi Han, Xuefu Dong, Yuuki Nishiyama, Kaoru Sezaki

2023 IEEE 24th International Symposium on a World of Wireless, Mobile and Multimedia Networks (WoWMoM)

Distracted riding behavior is one of the main causes of bicycle-related traffic accidents, resulting in a large number of casualties and economic losses every year. There is an urgent need to address this problem by accurately detecting distracted riding behaviors. Inspired by the observation that distracted riding behaviors induce unique head motion features that respond to the rider's attention, we present the HeadSense, a helmet-based system that not only monitors the visual search episode of the rider but also detects distracted riding behaviors. Specifically, HeadSense leverages the inertial motion unit (IMU) to recognize distracted behaviors such as using smartphones, attracting to the roadside element, and abreast riding. We designed, implemented, and evaluated HeadSense through extensive experiments. We conducted experiments with 19 participants inside the university's campus. The experimental results show that HeadSense can achieve an overall accuracy of 86.14% while monitoring visual search episodes. Moreover, HeadSense can detect the occurrence of distracted riding behaviors with an average precision of up to 85.04%.

Detecting Hand Hygienic Behaviors In-the-Wild Using a Microphone and Motion Sensor on a Smartwatch

Haoyu Zhuang, Liqiang Xu, Yuuki Nishiyama, Kaoru Sezaki

Distributed, Ambient and Pervasive Interactions

In recent years, the emergence of the COVID-19 pandemic has led to new viral variants, such as Omicron. These variants are more harmful and impose more restrictions on people’s daily hygiene habits. Therefore, during the COVID-19 pandemic, it is logical to automatically detect epidemic protective behaviors without user intent. In this study, we used multiple sensor data from an off-the-shelf smartwatch to detect several defined behaviors. To increase the utility and generalizability of the research results, we collected audio and inertial measurement unit (IMU) data from eight participants in real environments over a long period. In the model-building process, we first created a binary classification between hand hygiene behaviors(hand washing, disinfection, and face-touching) and daily behavior. Then, we distinguished between specific hand hygiene behaviors based on audio and IMU. Ultimately, our model achieves 93% classification accuracy for three behaviors(Hand washing, face touching, and disinfection). The results prove that the accuracy of the classification of behaviors has improved remarkably, which also emphasizes the feasibility of recognizing hand hygiene behaviors using inertial acoustic data.

HeadMon: Head Dynamics Enabled Riding Maneuver Prediction

Zengyi Han, Liqiang Xu, Xuefu Dong, Yuuki Nishiyama, Kaoru Sezaki

2023 IEEE International Conference on Pervasive Computing and Communications (PerCom)

Although micro-mobility brings convenience to modern cities, they also cause various social problems, such as traffic accidents, casualties, and substantial economic losses. Wearing protective equipment has become the primary recommendation for safe riding. However, passive protection cannot prevent the occurrence of accidents. Thus, timely predicting the rider's maneuver is essential for active protection and providing more time to avoid potential accidents from happening. Through the qualitative study, we argue that we can use the rider's head dynamic as an information source to predict the rider's following maneuvers. We accordingly present HeadMon, a riding maneuver prediction system for safe riding. HeadMon utilizes the head dynamics of a rider by installing an inertial measurement unit on the helmet. It uses the extracted head dynamics features as the input of the deep learning architecture to achieve prediction. We implemented the HeadMon prototype on Android smartphone as a proof of concept. Through comprehensive experiments with 20 participants, the result demonstrates the excellent performance of HeadMon: not only could it achieve an overall precision of at least 85% for maneuver prediction under a 4s prediction time gap, but it also could keep a high accuracy under a low sampling rate. The low-cost feature of HeadMon allows it to be readily deployable and towards more safety riding.

Preliminary Study for Classifying Baby Stroller-related Parenting using Smartphones

Zengyi Han, Xuefu Dong, Yuuki Nishiyama, Kaoru Sezaki

Activity and Behavior Computing

The stroller, as a necessary tool for parents' daily lives of infant care, is rich in information about babysitting-related, however, they are unexplored. Existing stroller studies usually focus on the hardware aspects such as automatic braking and self-propelling, leaving less attention on infant mobility. Nevertheless, such potential information might open up new perspectives to urban studies, physiology studies, and studies in other fields. Therefore, to extract the potential information from everyday stroller usage, we proposed the idea of leveraging ubiquitous devices such as smartphones to automatically monitor different stroller-related behaviors. Two built-in inertial measurement units (IMU) could enable a daily stroll-related interaction log, analysis, and eventually better parenting.

Detecting Face-Mask Wearing Status Using Motion Sensors in Commercially Available Smartwatches

Shota Ono, Yuuki Nishiyama, Kaoru Sezaki

2022 IEEE International Conference on E-health Networking, Application & Services (HealthCom)

Wearing a mask considerably mitigates the risk of infection from droplets. Automatic detection of whether a person wears a mask in his/her daily life and the type of masks the person wears can provide useful information for various services such as infection risk assessment, just-in-time alerts, and lifelogging. However, such automatic detection is difficult without the use of video processing or specialized equipment. In this study, the motion sensor of a commercially available smartwatch was used to detect the mask-wearing status. An investigation of the acceleration characteristic and an evaluation experiment of the mask-wearing state detection model revealed an accuracy of approximately 90% when specific motions were classified using motion sensors and machine learning. Furthermore, 98% accuracy was achieved when classifying sitting and walking activities.

Enhancing Self-Protection: What Influences Human’s Epidemic Prevention Behavior during the COVID-19 Pandemic

Liqiang Xu, Yuuki Nishiyama, Kaoru Sezaki

Distributed, Ambient and Pervasive Interactions. Smart Living, Learning, Well-Being and Health, Art and Creativity: 10th International Conference, DAPI 2022, Held as Part of the 24th HCI International Conference, HCII 2022, Virtual Event, June 26 – July 1, 2022, Proceedings, Part II

Under the circumstance of the rapid spread of the COVID-19 pandemic, enhancing human’s awareness of self-protection is one practical method to slow down the epidemic. In this study, we utilize mobile sensing to track human activity and guide human’s epidemic prevention behavior by gamified feedback techniques by our developed application. Virtually, human’s self-protection awareness is affected by many factors and the measures to enhance people’s self-protection behavior against the epidemic COVID-19 has always been an unresolved issue. In order to search for factors that influence human’s self-protection behavior, we analyzed the relationships between various human activities and the percentage complete of human’s self-protection behavior and we have extracted some more general conclusions from the results. Based on our data analysis results, we also made some proposals to enhance self-protection behavior. Meanwhile, our study illustrates the effectiveness of the method that analyzes human self-protection behavior through mobile sensing. Our study also validates the effectiveness of persuasive technology on human’s self-protection behavior against the COVID-19 pandemic and therefore we advocate enhancing human’s self-protection awareness through external intervention and guidance by smart device.

DoubleCheck: Single-Handed Cycling Detection with a Smartphone

Xuefu Dong, Zengyi Han, Yuuki Nishiyama, Kaoru Sezaki

2022 IEEE International Conference on Systems, Man, and Cybernetics (SMC)

Riding bicycles with only one hand on the handlebar can severely undermine the operator’s steering capability and threaten road and transportation safety. Prior studies have exploited motion sensors to detect riding contexts and recognize related behaviors. Nevertheless, they fail to integrate a scheme to account for single-handed riding with elements crucial to danger prevention: awareness of the surroundings, response to danger, and convenient adoption. In this work, we proposed, designed, and implemented DoubleCheck: a smartphone-based real-time framework for cycling hand detection and distraction recognition. The method monitors handlebar holding on different road surfaces and recognizes hazardous distraction activities related to single-handed cycling using motion signals captured by a built-in inertial measurement unit in a handlebar-borne smartphone. It was designed on the premise that single-handed cycling enabled operators to adapt their body movements to different (often distracting) activities. We conducted an evaluation experiment using 22 participants on asphalt and pavement. The results indicate that DoubleCheck achieves an F1-score of 0.96 for hand detection and 0.69 for distraction recognition, demonstrating its efficacy as a candidate rider-safety precautionary measure.

A Plug-in Memory Network for Trip Purpose Classification

Suxing Lyu, Tianyang Han, Yuuki Nishiyama, Kaoru Sezaki, Takahiko Kusakabe

Proceedings of the 30th International Conference on Advances in Geographic Information Systems

Trip purpose plays a critical role in reflecting human mobility behavior. However, it is relatively difficult to determine. With the rapid growth of urban mobility and big mobile data, utilizing these data for trip purpose classification has been a long-term objective to enhance travel demand and behavior models used in urban planning. Although studies on this topic have been extensively conducted, most past research preferred relying on traveler attributes or long-term travel histories to achieve accurate results. These data could be privacy sensitive and often do not satisfy real-world scenarios. This study addresses the problem of classifying trip purpose by only space activity information to avoid privacy conflict. 1) External memories are collected from factorized components based on the non-negative Tucker decomposition scheme. 2) These memories are extended by the cross-attention mechanism to achieve feature augmentation. 3) Subsequently, a novel concept called "latent mode alignment" is proposed. By leveraging the linear characteristics of external memories, geographic contextual latent modes are represented and matched with travel activities; this procedure is called älignment." 4) The gate mechanism controls the eventual outputs for update. The proposed plug-in memory network (PMN), combined with baseline models, effectively outperforms the original settings. Moreover, combination models are validated with strong tolerance through missing data tests, which are common and problematic in real-world scenarios. The proposed PMN is a plug-and-play design that is easy to combine with newly developed classification models, and other memory collection methods can be expected.

Estimation of Greenhouse Gas Emission Reduction from Shared Micromobility System

Helinyi Peng, Yuuki Nishiyama, Kaoru Sezaki

2021 IEEE Green Energy and Smart Systems Conference (IGESSC)

Shared micromobility is widely recognized as an environmentally friendly travel mode and a critical component of transportation decarbonization. However, quantitatively assessing its environmental impact using real-world trip data is an unresolved and challenging subject. In this research, we proposed a system combining machine learning algorithms and the Monte Carlo simulation to address this issue. First, several machine learning algorithms (Random Forest, XGBoost, and LightGBM) were utilized to identify citizens’ travel mode choice preferences and then estimate the substituted travel mode of each micromobility trip. Second, to ensure the reliability of the final environmental impact assessment, the Monte Carlo simulations were used to simulate the substituted mode of each trip. Then the environmental impacts were calculated based on the life cycle greenhouse gas emissions. Instead of estimating a specific number, we obtained a probabilistic outcome for environmental impacts by using the Monte Carlo simulation, which considers the uncertainty. According to the studies, the shared bike service and the shared e-scooter service have positive environmental impacts. However, these effects are limited compared to the transportation sector’s total emissions. The most compelling reason is that shared micromobility comprises a minuscule part of total urban travel.

An Unsupervised Learning-based Approach for User Mobility Analysis of E-Scooter Sharing Systems

Hidenaga Ushijima, Shunsuke Aoki, Peng Helinyi, Yuuki Nishiyama, Kaoru Sezaki

2021 IEEE International Intelligent Transportation Systems Conference (ITSC)

Human mobility analysis is a key method for understanding urban dynamics and mobility optimization. Novel last-mile mobility, called micromo-bilities, that includes shared bicycles, electric bicycles (e-bikes), and electric scooters are seeing rapid widespread acceptance in major cities. Compared with existing mobility data such as cars, buses, and trains, the majority trip distance of micromobilities is short, typically less than a few miles. The riders use them for commuting, sightseeing, shopping, and/or fun. By using the mobility data of micromobilities, we can observe more fine-grained human mobility in urban areas than existing data sources. In this paper, we present an unsupervised learning-based technique to analyze human mobility in urban areas and to study user clusters for such micromobility services. In our approach, we cluster user mobility patterns by using non-negative tensor factorization (NTF) from area-based trip data (which only included locations of origin and destination). Our approach was applied to micromobility data collected from Chicago and Washington, D.C., and we observed characteristic patterns.

A Run-time Dynamic Computation Offloading Strategy in Vehicular Edge Computing

Hong Duc Nguyen, Shunsuke Aoki, Yuuki Nishiyama, Kaoru Sezaki

2021 IEEE 94th Vehicular Technology Conference (VTC2021-Fall)

In vehicular edge computing (VEC), offloading the tasks to the nearby resource-rich edge servers helps each vehicle enhance computational capabilities and improve in-vehicle applications' performance. However, the concentration of travel at specific spaces and times poses significant challenges on the load-balancing and scheduling of computation tasks at the edge servers. This paper studies a low-complexity dynamic online offloading strategy that efficiently reduces task delay and computing resource consumption in the multi-user, multiserver vehicular edge computing scenarios. Our design addresses issues of computation task placement and execution order of the tasks on each server. We use a realistic approach that vehicles generate tasks over time, and the set of the tasks is unknown in advance so that the offloading decisions are made in runtime. Extensive simulations are conducted on a real mobility trace of Luxembourg city, and the results show that the proposed algorithm effectively improves the offloading utility of the system.

MiMoSense: An Open Crowdsensing Platform for Micro-Mobility

Zengyi Han, Hong Duc Nguyen, Shunsuke Aoki, Yuuki Nishiyama, Kaoru Sezaki

2021 IEEE International Conference on Intelligent Transportation (ITSC)

The use of micro-mobility (e.g., bicycle and scooter) and their data for urban sensing and rider assessment is becoming increasingly popular in research. However, different research topics require different sensor setups; no general data collecting tools for the micro-mobility makes the researcher who wishes to collect data has to build their own collecting system from scratch. To this end, we present MiMoSense, an open crowdsensing platform for micro-mobility. MiMoSense consists of two components: (1) MiMoSense server, which is set up on the cloud, and used to manage sensing studies and the collected data for research and sharing. (2) MiMoSense client, uses micro-mobility carrying various sensors and IoT devices to collect multiple kinds of data during traveling. As a reusable open-source software, MiMoSense shifts the researcher's focus from software development to sensing data analysis; it can help researchers quickly develop an extensible platform for collecting micro-mobility's raw sensing data and inferring traveling context. We have evaluated MiMoSense's battery consumption, message latency and discuss its use.

Towards Infectious Disease Risk Assessment in Taxis using Environmental Sensors

Hidenaga Ushijima, Shota Ono, Yuuki Nishiyama, Kaoru Sezaki

Distributed, Ambient and Pervasive Interactions

The spread of Coronavirus disease of 2019 (COVID-19) has reaffirmed the importance of ventilation in enclosed public spaces. Studies on air quality in public spaces such as classrooms, hospitals, and trains have been conducted in the past. However, the interior of a taxi, where an extremely small space is shared with an unspecified number of people, has not been sufficiently studied. This is a unique environment where ventilation is important. This study compared ventilation meth-ods focusing on the CO2 concentration in the cabin, and evaluated the frequency of ventilation in an actual taxi using sensing technology

iOS Crowd-Sensing Won't Hurt a Bit!: AWARE Framework and Sustainable Study Guideline for iOS Platform

Yuuki Nishiyama, Denzil Ferreira, Yusaku Eigen, Wataru Sasaki, Tadashi Okoshi, Jin Nakazawa, Anind K Dey, Kaoru Sezaki

Distributed, Ambient and Pervasive Interactions

The latest smartphones have advanced sensors that allow us to recognize human and environmental contexts. They operate primarily on Android and iOS, and can be used as sensing platforms for research in various fields owing to their ubiquity in society. Mobile sensing frameworks help to manage these sensors easily. However, Android and iOS are constructed following different policies, requiring developers and researchers to consider framework differences during research planning, application development, and data collection phases to ensure sustainable data collection. In particular, iOS imposes strict regulations on background data collection and application distribution. In this study, we design, implement, and evaluate a mobile sensing framework for iOS, namely AWARE-iOS, which is an iOS version of the AWARE Framework. Our performance evaluations and case studies measured over a duration of 288 h on four types of devices, show the risks of continuous data collection in the background and explore optimal practical sensor settings for improved data collection. Based on these results, we develop guidelines for sustainable data collection on iOS.

Challenges of Parkinson’s Disease: User Experiences with STOP

Elina Kuosmanen, Valerii Kan, Julio Vega, Aku Visuri, Yuuki Nishiyama, Anind K Dey, Simon Harper, Denzil Ferreira

Proceedings of the 21st International Conference on Human-Computer Interaction with Mobile Devices and Services

Parkinson's disease (PD) is the second most common neurodegenerative disorder, impacting an estimated seven to ten million people worldwide. Measuring the symptoms and progress of the disease, and medication effectiveness is currently performed using subjective measures and visual estimation. We developed and evaluated a mobile application, STOP for tracking hand's motor symptoms, and a medication journal for recording medication intake. We followed 13 PD patients from two countries for a 1-month long real-world deployment. We found that PD patients are willing to use digital tools, such as STOP, to track their medication intake and symptoms, and are also willing to share such data with their caregivers and medical personnel to improve their own care.

Mobile-Based Monitoring of Parkinson’s Disease

Elina Kuosmanen, Valerii Kan, Aku Visuri, Julio Vega, Yuuki Nishiyama, Anind K Dey, Simon Harper, Denzil Ferreira

Proceedings of the 17th International Conference on Mobile and Ubiquitous Multimedia

Towards health exercise behavior change for teams using life-logging

Yuuki Nishiyama, Tadashi Okoshi, Takuro Yonezawa, Jin Nakazawa, Kazunori Takashio, Hideyuki Tokuda

2014 IEEE 16th International Conference on e-Health Networking, Applications and Services (Healthcom)

Recent technological trends on mobile/wearable devices and sensors have been enabling increasing number of people to collect and store their “life-logs” easily in their daily lives. Beyond exercise behavior change of individual user, our research focus is on the behavior change of teams, based on life-logging technologies and information sharing. In this paper, we propose and evaluate six different types of information sharing model among team members for their exercise promotion, leveraging concepts of “competition” and “collaboration”. According to our experimental mobile web application for exercise promotion and extensive user study among 64 total users for three weeks, the model with “external competition” technique resulted the most effective performance for competitive teams such as sport teams.

A Mechanical Wi-Fi Antenna Device for Automatic Orientation Tuning with Bayesian Optimization

Akihito Taya, Yuuki Nishiyama, Kaoru Sezaki

2026 IEEE 23rd Annual Consumer Communications & Networking Conference (CCNC)

Toward Detecting Postpartum Depression Using Passive Mobile Sensing: Exploratory Analysis

Jia Tang, Xiuwen Gu, Akihito Taya, Kaoru Sezaki, Yuuki Nishiyama

Companion of the 2025 on ACM International Joint Conference on Pervasive and Ubiquitous Computing

Postpartum depression (PPD) is a prevalent mental health issue with severe consequences, yet its early detection is hampered by limitations in traditional screening methods like self-reported questionnaires. This study explores the potential of passive mobile sensing (PMS) to objectively assess PPD symptoms in postpartum women in Japan. Our iOS application, ZeroPPD, continuously collected sensor-derived behavioral metrics (e.g., mobility, activity, and phone usage) from 21 participants over four weeks. During this period, participants also submitted self-reported data on infant and maternal sleep, along with periodic Edinburgh Postnatal Depression Scale (EPDS) scores. Statistical analyses, including Linear Mixed-effects Models, identified significant correlations between specific behavioral features and EPDS scores. Key findings revealed specific behavioral patterns associated with PPD, including infant-related sleep disruptions and maternal caregiving stress (e.g., increased morning baby crying, longer nighttime stay durations), altered daytime activity and stationary behaviors (e.g., frequent afternoon activity bouts linked to lower EPDS), and potentially stress-related mobility. These results highlight PPD-specific digital biomarkers, suggesting that PMS can offer an effective pathway for continuous, objective monitoring of PPD-related behavioral changes, paving the way for the subsequent development of data-driven early risk identification tools and timely interventions.

Recognizing Hidden-in-the-Ear Private Key for Reliable Silent Speech Interface Using Multi-Task Learning

Xuefu Dong, Liqiang Xu, Lixing He, Zengyi Han, Kenneth Christofferson, Yifei Chen, Akihito Taya, Yuuki Nishiyama, Kaoru Sezaki

Companion of the 2025 on ACM International Joint Conference on Pervasive and Ubiquitous Computing

Silent speech interaction (SSI) enables hands-free input without audible vocalization, but most SSI systems do not verify speaker identity. We present HEar-ID, which uses consumer active noise–canceling earbuds to capture low-frequency ''whisper'' audio and high-frequency ultrasonic reflections. Features from both streams pass through a shared encoder, producing embeddings that feed a contrastive branch for user authentication and an SSI head for silent spelling recognition. This design supports decoding of 50 words while reliably rejecting impostors, all on commodity earbuds with a single model. Experiments demonstrate that HEar-ID achieves strong spelling accuracy and robust authentication.

AEDHunter: Investigating and Enhancing AED Retrieval through a Mobile Application

Helinyi Peng, Akihito Taya, Yuuki Nishiyama, Kaoru Sezaki

Proceedings of the Extended Abstracts of the CHI Conference on Human Factors in Computing Systems

Early defibrillation improves survival rates in out-of-hospital cardiac arrest (OHCA), yet public use of automated external defibrillators (AEDs) remains low due to unawareness of their locations. Existing solutions like 2D maps are often insufficient during emergencies, especially indoors where navigation is challenging. We developed AEDHunter, a gamified, location-based mobile application that enhances public awareness of AED locations by transforming the retrieval process into an engaging adventure. AEDHunter leverages smartphone sensors to analyze user retrieval patterns and navigation behaviors in real-world environments. Through experiments involving repeated interactions with AEDHunter, we investigate how users’ confidence, willingness, and behaviors in locating AEDs evolve. We also introduce a semi-supervised classification method to detect mobile device usage during these tasks, laying the groundwork for analyzing nuanced AED retrieval behaviors. This work-in-progress provides initial insights into enhancing AED awareness and retrieval through gamified applications.

The Bidirectional Relationship between Emotional Change and Physical Movement Activities: An Analysis Using Propensity Score Matching Methods

Zeyu Wu, Yuuki Nishiyama, Yuya Shibuya

2024 IEEE International Conference on Big Data (BigData)

This study investigates the bidirectional relationship between emotional changes and physical activity, specifically focusing on the number of steps taken. Using Propensity Score Matching (PSM), we analyzed how fluctuations in emotional states influence physical activity and, conversely, how increases/decreases in daily steps impact subsequent emotional well-being. This study uses a dataset containing data on daily steps and emotional status collected via a smartphone application (N=123). Our findings indicate that increases in the number of steps significantly increase the subsequent emotion report positively. Additionally, changes in emotional status have relations with a subsequent number of steps. These results suggest a reciprocal influence between emotional and physical activities, highlighting the importance of integrating physical and mental health interventions.

Investigating Acceptable Voice-based Notification Timings through Earable Devices: A Preliminary Field Study

Manaka Ito, Kota Tsubouchi, Nobuhiko Nishio, Masamichi Shimosaka, Akihito Taya, Kaoru Sezaki, Yuuki Nishiyama

Companion of the 2024 on ACM International Joint Conference on Pervasive and Ubiquitous Computing

Earable devices, a subset of wearable technology, are designed to be worn on the ear and used in daily life. These innovative devices enable users to receive voice-based notifications through a minute built-in speaker without requiring any user operations, seamlessly integrating technology into everyday activities. The timing deemed acceptable for receiving voice-based notifications through earable devices varies based on the user and surrounding situation; thus, inappropriate notification timing may reduce usability. However, determining the safest and most comfortable timing for voice-based notifications using earable devices remains unclear. This study investigates the acceptable timing of voice-based notifications through earable devices. To explore the acceptable timing, we developed a smartphone application, SoNotify, which can send dummy voice-based notifications and collect sensor data on a smartphone and an earable device. Our field studies with eight participants showed that voice-based notifications were highly acceptable during outdoor walking, with an acceptance rate of approximately 86%. However, users tended to refuse notifications in situations in which they needed to concentrate on avoiding collisions with pedestrians, cyclists, or vehicles.

Toward Detecting Student-Athletes' Condition Using Passive Mobile and Wearable Sensing

Zhenbo Wang, Akihito Taya, Takaaki Kato, Kaoru Sezaki, Yuuki Nishiyama

Companion of the 2024 on ACM International Joint Conference on Pervasive and Ubiquitous Computing

Student-athletes (SAs) face stress from balancing athletic and academic demands; therefore, monitoring their physical and mental stress is crucial to ensure better well-being. Questionnaires and dedicated measurement equipment have been used to assess the mental and physical status of SAs. However, these methods are not scalable and are difficult to use continuously. In this paper, we propose a method for monitoring the physical and mental state of SAs using passive mobile and wearable sensing technology to minimize the burden associated with monitoring. First, we developed a platform for collecting daily, training, and resetting behavior data from smartphones and wearable devices. Second, as a preliminary study, we collected the behavior and Stress and Recovery States Scale (SRSS) data for four weeks from 19 SAs and analyzed the collected data to understand their unique behavior patterns and living environments. The results demonstrate that wearable devices and smartphones can automatically collect data on "exercise intensity during competitions" and "lifestyle patterns," specifically tailored to the mental and physical states of SAs. Furthermore, this preliminary research establishes a foundation for future efforts using machine learning to predict the physical and mental states of SAs.

Toward Detecting Maternity Neurosis by Using Passive Mobile Sensing: Preliminary Investigation

Xiuwen Gu, Akihito Taya, Yuuki Nishiyama, Kaoru Sezaki

2024 IEEE International Conference on E-health Networking, Application & Services (HealthCom)

Child-rearing depression, triggered by the chronic stress of parenting, can lead to serious mental health issues if not detected early. This study uses passive mobile sensing to analyze the behavioral and psychological patterns of households with preschool children. By collecting data from 131 participants (including 18 parents of preschoolers), we aim to differentiate child-rearing anxiety and behavior patterns. Our focus includes step counts, location data, call frequency, and psychological states. Results indicate that parents of preschoolers have fewer steps, visit fewer locations, and have higher call activity. They also show higher stress and anxiety but lower depression levels, suggesting that family support may mitigate depressive symptoms. These insights could aid in developing early detection and intervention strategies for child-rearing depression.

Poster: Location Awareness in AED Retrieval: An Simulation-Based Investigation

Helinyi Peng, Akihito Taya, Yuuki Nishiyama, Kaoru Sezaki

Proceedings of the 22nd Annual International Conference on Mobile Systems, Applications and Services

Public awareness of automated external defibrillator (AED) locations is crucial for prompt retrieval in cardiac emergencies. We propose a simulation-based approach as a preliminary step towards developing gamified mobile apps to enhance this awareness. By simulating AED retrieval in real-world pedestrian networks under various scenarios, we identify key elements that can improve retrieval efficiency. Our findings confirm the viability of the framework and highlight crucial aspects for improvement towards efficient future applications.

Poster: Towards Estimating UV Index with a Smartphone Utilizing GNSS Signals as a Point Cloud

Subaru Atsumi, Riku Ishioka, Kota Tsubouchi, Yuuki Nishiyama, Kaoru Sezaki

Proceedings of the 22nd Annual International Conference on Mobile Systems, Applications and Services

Monitoring and controlling the exposure of an individual to ultraviolet (UV) radiation is crucial for personal health. The use of the global navigation satellite system (GNSS) signals received by a personal off-the-shelf smartphone has been studied as a novel estimation method. In the existing method, satellites are grouped based on their positions and the signal information is represented by group statistics, leading to a coarse estimation. We propose a new UV index estimation method that directly utilizes satellite-wise information and their spatial relationships with a point-cloud neural network, considering the similarity between GNSS signals and point clouds. We collected GNSS signals and UV index data from two locations within the same area and demonstrated that the proposed method enhances the estimation accuracy and smoothness.

Exploiting Spatial and Descriptive Information for Generative Compression

Eri Hosonuma, Taku Yamazaki, Takumi Miyoshi, Akihito Taya, Yuuki Nishiyama, Kaoru Sezaki

2024 IEEE 21st Consumer Communications & Networking Conference (CCNC)

Experimental Evaluation Toward Mobility-Driven Model Integration Between Edges

Shota Ono, Taku Yamazaki, Takumi Miyoshi, Akihito Taya, Yuuki Nishiyama, Kaoru Sezaki

2024 IEEE 21st Consumer Communications & Networking Conference (CCNC)

A Trajectory Estimation Method from Spatially Sparse and Noisy Beacon Data Based on Spring Dynamics

Junya Maruyama, Yudai Honma, Yuuki Nishiyama, Yoshihiro Kawahara

Proceedings of the 10th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation

Analysis of trajectory data within buildings offers insights for optimizing environmental design and habitability. However, data from indoor location sensors tend to be sparse and noisy. This makes it difficult for conventional route estimation models to be applied effectively. Our study seeks to derive detailed, temporally, and spatially rich trajectory data from this compromised sensor information. We achieve this by interpreting trajectories as continuous stay points. To facilitate this, we introduce a building corridor network that conceptualizes buildings as a series of points. Routes are inferred using a sequence estimation model applied to this network. This approach employs spring dynamics, which balance the resistance to staying with the attraction to specific beacons, via mathematical optimization. Notably, our model can deduce a trajectory of 131 points from only 15 beacons with, an accuracy rate of 87. Our method presents a promising avenue for capturing extensive route data.

Assessing Infant and Toddler Behaviors through Wearable Inertial Sensors: A Preliminary Investigation

Ayaka Onodera, Riku Ishioka, Yuuki Nishiyama, Kaoru Sezaki

Companion Publication of the 25th International Conference on Multimodal Interaction

Accurately assessing a child’s health status and taking appropriate action in case of illness or other emergencies is crucial to recording and sharing a childcare situation, such as sleeping hours, amount of exercise, and timing of meals. Although numerous applications and systems have been proposed to assist in recording and sharing these records, the process is still performed manually, representing a significant burden for parents. Therefore, automatic recording of infants’ and toddlers’ daily activities is required. Moreover, existing automatic infant behavior recognition methods have significant limitations on the space and target of application. In this study, we implement a machine learning model to investigate and propose a method to classify typical daily behaviors of infants and toddlers using a chest-mounted low-sampling rate accelerometer. In particular, the proposed method classifies eight activities: sleeping, crawling, walking, standing, sitting, drinking milk, eating baby food, and holding by a caregiver. As a dataset, we collected accelerometer data for nearly 18 hours with reference videos from ten infants and toddlers between 6 and 24 months. Based on the data, we extracted 60 time- and frequency-domain features calculated from the single accelerometer and a user feature to recognize the target daily activities. The performance evaluation considering different window sizes and machine learning models showed that our classification model reaches nearly 80% accuracy in the best-case scenario.

Smartwatch-Based Sensing Framework for Continuous Data Collection: Design and Implementation

Yuuki Nishiyama, Kaoru Sezaki

Adjunct Proceedings of the 2023 ACM International Joint Conference on Pervasive and Ubiquitous Computing & the 2023 ACM International Symposium on Wearable Computing

Smartwatches are an increasingly popular technology that employs advanced sensors (e.g., location, motion, and microphone) comparable to those used by smartphones. Passive mobile sensing, a method of acquiring human behavior data from mobile and wearable devices inconspicuously, is widely used in research fields related to behavior analysis. In combination with machine learning, passive mobile sensing can be used to interpret various human and environmental contexts without requiring user intervention. Because smartwatches are always worn on the wrist, they have the potential to collect data that cannot be collected by smartphones. However, the effective use of smartwatches as platforms for passive mobile sensing poses challenges in terms of battery life, storage, and communication. To address these challenges, we designed and implemented a tailored framework for off-the-shelf smartwatches. We evaluated power consumption under eight different sensing conditions using three smartwatches. The results demonstrate that the framework can collect sensor data with a battery life of 16-31 h depending on the settings. Finally, we considered potential future solutions for optimizing power consumption in passive sensing with off-the-shelf smartwatches.

Cooperative Local Distributed Machine Learning Considering Communication Latency and Power Consumption

Shota Ono, Taku Yamazaki, Takumi Miyoshi, Yuuki Nishiyama, Kaoru Sezaki

2023 IEEE 20th Annual Consumer Communications & Networking Conference (IEEE CCNC 2023)

Enabling Block Transmission on Backoff-based Opportunistic Routing

Eri Hosonuma, Yuuki Nishiyama, Kaoru Sezaki, Takumi Miyoshi, Taku Yamazaki

2023 IEEE 20th Annual Consumer Communications & Networking Conference (IEEE CCNC 2023)

Opportunistic Division and Allocation of Machine Learning Task for WSN

Eri Hosonuma, Nobuyuki Tanaka, Takuma Yamazaki, Akihito Taya, Yuuki Nishiyama, Kaoru Sezaki, Takumi Miyoshi, Taku Yamazaki

2023 Fourteenth International Conference on Ubiquitous and Future Networks (ICUFN)

Poster abstract: Room Scale Localization Improvement Utilizing Stay Time Characteristics of Each Room

Ryoto Suzuki, Yuuki Nishiyama, Hiroaki Murakami, Yoshihiro Kawahara, Kaoru Sezaki

Proceedings of the 20th Conference on Embedded Networked Sensor Systems, Boston, USA

Indoor localization technology is one of the most important topics in the fields of Internet of Things (IoT) and ubiquitous computing, and it has attracted much attention in recent years due to its ability to enable a variety of services. Localization methods using Bluetooth or Wi-Fi signal strength can introduce a low-cost location estimation system. However, due to the instability of the received signal strength and signals leaking from adjacent rooms, a simple method based on signal strength alone frequently results in misjudgment, depending on the signal propagation characteristics. In this paper, we propose a method to suppress misjudgment by considering the characteristics of stay time in different rooms. Our proposed method estimates the user state by fitting the distribution of time spent in each room to a Weibull distribution and applying survival analysis. The experimental results suggest that the method will provide more accurate information about the rooms in which users stay.

Poster abstract: UV index estimation leveraging GNSS sensors on smartphones

Riku Ishioka, Yuuki Nishiyama, Kota Tsubouchi, Kaoru Sezaki

Proceedings of the 20th Conference on Embedded Networked Sensor Systems, Boston, USA

Monitoring the amount of UV irradiance to which individuals are exposed and ensuring that every individual receives the optimal amount has been the subject of extensive research. In previous research, the UV index was estimated using cell phone cameras, light sensors on smartphones, or wearable UV sensors. We propose a method for estimating the UV index using the widespread global navigation satellite system (GNSS) sensors available on smart-phones. In contrast to approaches that require the sensor to be exposed continuously to the irradiance, this method, which leverages GNSS sensors, has the potential advantage of enabling UV index measurement simply by carrying the phone as usual. As a first step in measuring the index using GNSS sensors, GNSS data were collected from cell phones placed at three locations in a single area; the OpenUV API was utilized as a baseline. The proposed method achieved a mean absolute error of 0.1523, which significantly outperformed the baseline.

Poster abstract: Recursive Queueing Estimation Using Smartphone-based Acoustic Ranging

Issey Sukeda, Hiroaki Murakami, Yuuki Nishiyama, Hiroaki Murakami, Yoshihiro Kawahara

Proceedings of the 20th Conference on Embedded Networked Sensor Systems, Boston, USA

When customers wait their turn to place an order at a street vendor, they often form a spontaneous queue. Due to the presence of passers-by and people standing outside the queue, it is more difficult than one might think to distinguish between those in the queue and those not in the queue. In this paper, we consider a method that uses acoustic ranging to autonomously detect who is in line and in which order, under the condition that all customers have smartphones. The proposed method is unique in that it can distinguish whether a newly arrived user has joined the end of the queue or not by taking cues from the geometric properties of the queue. Our preparatory queueing simulations confirm that 92.5% of the queuers are estimated correctly.

Poster abstract: Convolutional Compressed Sensing for Smartphone Acceleration Data Compression

Liqiang Xu, Yuuki Nishiyama, Masamichi Shimosaka, Kota Tsubouchi, Kaoru Sezaki

Proceedings of the 20th Conference on Embedded Networked Sensor Systems, Boston, USA

As intelligent sensing and smartphone technologies have progressed, a huge amount of highly heterogeneous data have come to be stored in smartphones and uploaded to servers for analysis on a daily basis. This has led to vast storage overheads for users and companies. Hence, data compression becomes the most efficient strategy for suppressing the increase in storage overhead. Compressed sensing (CS) technology is one approach to compressing data, but traditional CS-based algorithms are significantly time-consuming and have low reconstruction performance. In light of these drawbacks, this paper proposes a compressed sensing framework that instead takes advantage of the low time cost and adaptive learning capability of deep learning methods, wherein a convolutional neural network (CNN) is used for compressing and reconstructing acceleration data. Our experiments with actual smartphone acceleration data show that the proposed method dramatically improves the reconstruction performance with very little reconstruction time compared with traditional compressed sensing methods.

MOCHA: Mobile Check-in Application for University Campuses Beyond COVID-19

Yuuki Nishiyama, Hiroaki Murakami, Ryoto Suzuki, Kazusato Oko, Issey Sukeda, Kaoru Sezaki, Yoshihiro Kawahara

The Twenty-third International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing

Users and operators of shared spaces must ensure safety in such areas to prevent the spread of COVID-19. Although each organization has operated a variety of safety-related systems, including contact tracing, congestion monitoring, and check-in services, it is unclear what elements, such as privacy protection level, benefits, and permission procedures, have promoted the usage of these systems. In this study, we created MOCHA, a platform for sharing and tracking room-level locations. This platform automatically detects visited places by scanning Bluetooth beacons in each room using smartphones and shares location data according to predefined user settings. The collected data is used for room-level contact tracing, congestion monitoring, and reservation services. According to >6,500 users' usage data for a year in a university, outlining the advantages of utilizing the app encouraged people to install the app, and reinforced connections in small private groups are encouraged to use the app continuously.

A Micro-mobility Sensing System to Portray Riding Styles

Zengyi Han, Yuuki Nishiyama, Kaoru Sezaki

Adjunct Proceedings of the 2022 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2022 ACM International Symposium on Wearable Computers

Riding style concerns the way a micro-mobility rider chooses to ride, and it plays a significant role in traffic safety. Portraying a rider’s riding style is a useful way to guide them towards safer riding behaviors, and offer fine-grained information for insurance companies and bike-sharing companies to provide better services. To this end, we propose a micro-mobility sensing system to portray riding styles. Utilizing helmet-mounted and bicycle handle-mounted inertial sensors, our sensing system is able to monitor the micro-mobility movement status, detect the maneuver behaviors, and the safety check condition of riders. With 10 participants’ experimental data, we present the feasibility of detecting maneuver behavior and recording the data that characterize the rider’s riding style with our sensing system, and therefore make it a viable addition to the micro-mobility.

A Preliminary Study for Detecting Visual Search Behaviors During Street Walking Using Earable Device

Kazuki Shimojo, Zengyi Han, Yuuki Nishiyama, Kaoru Sezaki

Adjunct Proceedings of the 2022 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2022 ACM International Symposium on Wearable Computers

Map applications on smartphones are powerful navigation tools for walking among places to visit for the first time and are used widely. On the other hand, checking the application tend to cause trouble on the road such as collisions with people, cars, and objects. To prevent the troubles, we need to detect the walker’s context regarding visual search behaviors and provide appropriate navigation information for the walker. In this paper, we propose a method to detect a walker’s context regarding visual search behaviors by using motion sensors on an earable device. We collected and investigated motion and gaze data from an earable device and gaze tracker respectively during street walking from five participants. Based on the investigation, we create a machine learning model for detecting looking around, smartphone, or normal during walking and stopping conditions. Our evaluations show that our models can detect more than 95% of walking and stopping conditions, and 71% of three details conditions during walking, respectively

Toward Measuring Conversation Duration Using a Wristwatch-type Wearable Device

Yuki Komatsu, Kazuki Shimojo, Yuuki Nishiyama, Kaoru Sezaki

2022 IEEE International Conference on Smart Computing (SMARTCOMP)

The frequency and duration of social contact, represented by conversation, is positively correlated with our physical and mental health. Therefore, a method that automatically measures social contact can provide insight into people's health conditions and risks. Even though off-the-shelf wristwatch-type wearable devices are widely used in our daily lives and have rich computational resources, they have not been used as a social-contact monitoring tool in everyday conditions. In this study, we propose a system, called Ohanashi, for continuously monitoring conversational events as a means of social contact in daily life, by edge processing on an off-the-shelf smartwatch. To monitor the conversational event, we developed an audio classification model and implemented it as a smartwatch application, which can classify conversation and noise from an audio stream. Our performance evaluation shows that the classification model can classify conversation and noise with more than 86% accuracy in both silent and noisy environments, and the system can monitor conversation events for more than 15 hours on a smartwatch.

Detecting Childcare Activities Using an Off-the-shelf Smartwatch

Yuki Kasahara, Yuuki Nishiyama, Kaoru Sezaki

2022 IEEE International Conference on Smart Computing (SMARTCOMP)

The childcare environment has significantly changed, owing to accelerating women's social advancement and the increasing number of nuclear families. Improving and supporting childcare have become major challenges in current society. Automatically recording and subsequently observing childcare activities can be used for various purposes to support childcare. However, methods to detect childcare activities using off-the-shelf devices have not yet been proposed. This study develops a method to detect childcare activities that parents perform for their babies using an off-the-shelf wearable device. We define nine childcare activities and develop corresponding detection models based on motion-sensor data from a smartwatch. Our evaluation in a laboratory setting resulted in classification performances of 71% (F1: 0.66).

DoubleCheck: Detecting Single-Hand Cycling with Inertial Measurement Unit of Smartphone

Xuefu Dong, Zengyi Han, Yuuki Nishiyama, Kaoru Sezaki

IEEE International Conference on Pervasive Computing and Communications (PerCom)

Riding bikes with only one hand on the handlebar can severely undermine the steering capability of riders and risk road safety. In this study, we propose a first detection framework for monitoring single-hand cycling on bicycle travel, called DoubleCheck. It is based on the premise that riders adapt their body movement during single-hand cycling, which is distinguishable to the sensors even amid noise from the exasperate road surface. The system can detect handlebar-holding under different road conditions using motion signals from a built-in inertial measurement unit (IMU) in a handlebar-mounted smartphone. We implemented the system and invited 10 participants for our evaluation experiment. Our results show that DoubleCheck achieved an F1-score of 0.94 for hand detection, proving its efficacy for real-life implementation to improve road safety.

Head Dynamics Enabled Riding Maneuver Prediction

Zengyi Han, Xuefu Dong, Yuuki Nishiyama, Kaoru Sezaki

Proceedings of the 20th Annual International Conference on Mobile Systems, Applications and Services

While micro-mobility brings convenience to the modern city, they also cause various social problems such as traffic accidents, casualties, and huge economic losses. Wearing protective equipment has become the primary recommendation for safe riding, but passive protection cannot prevent accidents from happening after all. Thus, timely predicting the rider's maneuver is essential for more active protection and buying more time to avoid potential accidents from happening. In this poster, we explore the feasibility of using riders' head dynamics to predict their riding maneuvers. Through ten participants' preliminary study, we observed that not only do riders' head movements appear ahead of their maneuvers but also head movement patterns are distinct with different maneuver intentions. We then construct a deep learning network using Long Short Term Memory, achieving 89% of accuracy on maneuver prediction.

Detecting Single-Hand Riding with Integrated Accelerometer and Gyroscope of Smartphone

Xuefu Dong, Zengyi Han, Yuuki Nishiyama, Kaoru Sezaki

Adjunct Proceedings of the 2021 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2021 ACM International Symposium on Wearable Computers

Single-hand cycling poses a safety threat with the decrement of riders’ response capacity. Recognizing risky behavior by prevalently used smartphones could lead to enhanced riding safety. In this work, we propose a single-hand cycling recognition method based on motion data acquired from the three-axis accelerometer and gyroscope integrated into a handlebar-installed smartphone. We conducted a 4-person experiment. The data result demonstrates that motion data of double-hand cycling clearly distinguishes from that of single-hand, revealing the chance to materialize a robust detection tool in smartphones to enable safer biking. For future work, we prepare to redesign the experiment under more sophisticated circumstances with an improved platform, thus scaling this sensing method for real-life usage.

Experience Sampling Tool for Repetitive Skills Training in Sports Using Voice User Interface

Yuuki Nishiyama, Kaoru Sezaki

Adjunct Proceedings of the 2021 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2021 ACM International Symposium on Wearable Computers

Repetitive skills training (RST) is a commonly used method for improving skills. Although wearable devices and existing context-aware technologies allow us to easily detect objective data during RST, subjective data have not been collected effectively, even though both objective and subjective data are important for RST. In this paper, we propose and implement a prototype system, called MiQ, to collect subjective data with minimum workload during RST for sports. MiQ allows us to record subjective data hands-free via a voice user interface (VUI). We also discuss the future scope of the proposed prototype system.

HeadSense: A Head Movement Detecting System for Micro-Mobility Riders

Zengyi Han, Xuefu Dong, Yuuki Nishiyama, Kaoru Sezaki

Adjunct Proceedings of the 2021 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2021 ACM International Symposium on Wearable Computers

Head movement for traffic visual searching, is one of the important factors in traffic safety. In this paper, we present the design, implementation, and preliminary evaluation of the HeadSense, a helmet device that detects the head movement of micro-mobility rider. HeadSense is capable of generating data streams using the embedded 9-axis inertial measurement unit (IMU) sensor. After the process of segmentation and classification algorithm, HeadSense can automatically detect an individual’s head movement sequence and visual search episodes, across the rider’s entire riding journey. Experiments with 5 participants show that our system achieves 94.7% for per-second level detection and 80.59% F1-score for per-episode level detection.

Towards Estimating UV Exposure Using GPS Signal Strength from a Carrying Smartphone

Soichiro Higuma, Kosuke Hatai, Yuuki Nishiyama, Kaoru Sezaki

2021 IEEE International Conference on Smart Computing (SMARTCOMP)

Owing to lifestyle changes, urbanization, and the COVID-19 pandemic, many people spend more time indoors and tend to receive less direct sunlight than before. Although excessive or inadequate ultraviolet (UV) exposure can be harmful to our physical and mental health, moderate UV exposure is essential for vitamin D (VD) production in the body. In this study, we estimate the UV exposure using an off-the-shelf smartphone, and explore the relationship between the UV values and GPS signal strength (C/N0). The results demonstrate that a strong correlation (R 2 = 0.73) between the UV values and carrier to noise density (C/N0) even if the smartphone and UV sensor are moved. Therefore, it is possible to estimate the UV exposure to some extent from a person's location, even while carrying a smartphone.

SelfGuard: Semi-Automated Activity Tracking for Enhancing Self-Protection against the COVID-19 Pandemic

Yuuki Nishiyama, Takuro Yonezawa, Kaoru Sezaki

Proceedings of the 18th Conference on Embedded Networked Sensor Systems, Virtual Event, Japan

Contagious diseases like COVID-19 spread periodically and threaten our lives. Self-protection, such as washing hands, wearing a mask, and staying home, are simple and practical solutions to safeguard against these diseases. Most governments and health departments recommend that people maintain self-protection. Although continuous self-protection effectively prevents the spread of infection, only the intent to self-protect is unsustainable in the long term. In this study, we design, develop, and deploy an application to track users' daily activities semi-automatically and enhance self-protection behavior using mobile sensing and gamified feedback techniques. Currently, more than 324 people have installed the app via AppStore, and 52 users have shared their activity data to our research group.

Towards Estimating UV Light Intensity Using GPS Signal Strength

Soichiro Higuma, Yuuki Nishiyama, Kaoru Sezaki

Adjunct Proceedings of the 2020 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2020 ACM International Symposium on Wearable Computers

Due to recent urbanization and changing lifestyles, many people have been spending more time indoors. Hence, they tend to receive less direct sunlight than ever before. Although excessive/inadequate UV exposure can be harmful to human health leading to illnesses such as skin cancer, spots, or depression, moderate UV exposure is necessary for vitamin D production in the body. Therefore, estimating UV exposure with a commonly used device is useful for maintaining a healthy lifestyle from excessive/inadequate UV exposure in our daily life. In this study, we aim to estimate UV exposure, and to this end, we used the GPS signal strength (C/No) collected from an off-the-shelf smartphone for exploring the relationship between UV values and C/No. We conducted an experiment and measured UV values and C/No from 10 places in two different situations. From the results, we observed a significant correlation (R2 more than 0.87) between UV values and C/No when all the data were divided by the sun/shade condition. This result supports the fact that UV values potentially can be inferred from C/No to some degree if the sun/shade condition can be detected.

Using IOS for Inconspicuous Data Collection: A Real-World Assessment

Yuuki Nishiyama, Denzil Ferreira, Wataru Sasaki, Tadashi Okoshi, Jin Nakazawa, Anind K Dey, Kaoru Sezaki

Adjunct Proceedings of the 2020 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2020 ACM International Symposium on Wearable Computers

Mobile Crowd Sensing (MCS) is a method for collecting multiple sensor data from distributed mobile devices for understanding social and behavioral phenomena. The method requires collecting the sensor data 24/7, ideally inconspicuously to minimize bias. Although several MCS tools for collecting the sensor data from an off-the-shelf smartphone are proposed and evaluated under controlled conditions as a benchmark, the performance in a practical sensing study condition is scarce, especially on iOS. In this paper, we assess the data collection quality of AWARE iOS, installed on off-the-shelf iOS smartphones with 9 participants for a week. Our analysis shows that more than 97% of sensor data, provided by hardware sensors (i.e., accelerometer, location, and pedometer sensor), is successfully collected in real-world conditions, unless a user explicitly quits our data collection application.

Using Mobile Sensing Technology for Capturing People Mobility Information

Chenwei Song, Masaki Ito, Yuuki Nishiyama, Kaoru Sezaki

Proceedings of the 3rd ACM SIGSPATIAL International Workshop on Prediction of Human Mobility

Capturing Subjective Time as Context and It’s Applications (Poster)

Takuro Yonezawa, Yuuki Nishiyama, Kei Hiroi, Nobuo Kawaguchi

Proceedings of the 17th Annual International Conference on Mobile Systems, Applications, and Services

We propose an integrated framework for sensing, recognizing and utilizing of subjective time as context. Various studies on experimental psychology have showed several factors which affects subjective time. Those factors should be partially captured by ubiquitous sensors such as smartphones and wearable devices, therefore, we tackle to create common and individual model for subjective time based on the sensor data. We report our first prototype implementation for the framework based on AWARE framework with adding experience sampling method for subjective time recognition. In addition, we discuss potential applications which leveraging advantages of subjective time as context.

Fish Species Annotation System for Environmental Detection by Marine Life as a Sensor

Naohiro Isokawa, Wataru Sasaki, Yuuki Nishiyama, Tadashi Okoshi, Jin Nakazawa

HCI Outdoors: A CHI 2018 Workshop on Understanding Human-Computer Interaction in the Outdoor, Canada, Montreal

The ocean was born on the earth 4.6 billion years ago. Life was born in the ocean, further evolved creatures created oxygen in the ocean, and that oxygen also changed the atmospheric composition of the Earth. From ancient times life has received many benefits from the ocean. Even in recent years, fluctuations in the marine environment are drawing attention as one of the major environmental problems. Because there is a possibility that it will have a big influence on human society and ecosystem. Therefore, in the field of oceanography, research to detect and predict changes in the marine environment in various ways is actively conducted. We aim to detect changes in the marine environment from the behavior of marine organisms (mainly fish). In this paper, we implemented a system for annotation of fish species which is an important module for our research. This system aims to collect accurate data on fish species by targeting professional oceanographers. For motivation for annotation work, the system provides analytical results and generation models as rewards for work. We also conducted a questionnaire survey of 14 professors, researchers, and students specializing in oceanography for annotation system.

Meal Photo SNS with Mutual Healthiness Evaluation for Improving Users’ Eating Habits

Yusaku Eigen, Yuuki Nishiyama, Takuro Yonezawa, Tadashi Okoshi, Jin Nakazawa

Proceedings of the 2018 ACM International Joint Conference and 2018 International Symposium on Pervasive and Ubiquitous Computing and Wearable Computers

We propose a Meal Photo SNS (called HealthyStadium) for improving users' eating habits by mutually assessing each others' health. This application's method is to evaluates pictures of meals leads to the realization of sustainable journaling. In addition, we implement competitive awareness that motivates the users to improve their eating habits by allowing users to share their healthiness ranking amongst the users. Through our study, we confirmed that our system enables users to motivate for improving eating habits. Also we found the number of records is increased by our system.

Senbay: A Platform for Instantly Capturing, Integrating, and Restreaming of Synchronized Multiple Sensor-Data Stream

Yuuki Nishiyama, Anind K Dey, Denzil Ferreira, Takuro Yonezawa, Jin Nakazawa

Proceedings of the 20th International Conference on Human-Computer Interaction with Mobile Devices and Services Adjunct

The spread of smartphones allows us to freely capture video and diverse hardware sensors' data (e.g., accel erometer, gyroscope). While recording such data is relatively simple, it is often challenging to share and restream this data to other people and applications. Such capability is very valuable for a range of applications such as a context-aware prototyping/developing platform, an integrated data recording and analysis tool, and a sensor-data based video editing system. To enable such complex operations, we propose Senbay, a platform for instant sensing, integrating, and restreaming multiple-sensor data streams. The platform embeds collected sensor data into a video frame using an animated two-dimensional barcode via real-time video processing. The video-embedded sensor data, dubbed Senbay Video, can be easily restreamed to other people and reused by data rich, context-aware applications.

Poster: Extensive Evaluation of Emotional Contagion on Smiling Selfies over Social Network

Wataru Sasaki, Mikio Obuchi, Kazuki Egashira, Naohiro Isokawa, Yuki Furukawa, Yuuki Nishiyama, Tadashi Okoshi, Jin Nakazawa

Proceedings of the 15th Annual International Conference on Mobile Systems, Applications, and Services

We propose "SmileWave", the first selfie social networking service to reveal the existence of emotional cognation through smiling selfies on the social network. We conducted multiple rounds of in-the-wild user studies with 86 cumulative total users for total duration of 5 weeks. Throughout the entire study, we confirmed the occurrence of smile-based emotional contagion over social network, not only in the momentary duration but in longer term period.

Poster Abstract: SmileWave - Sensing and Analysis of Smile-Based Emotional Contagion over Social Network

Wataru Sasaki, Yuki Furukawa, Yuuki Nishiyama, Tadashi Okoshi, Jin Nakazawa, Hideyuki Tokuda

2016 15th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN)

This paper proposes ''SmileWave", a system for revealing smile-based emotional contagion, propagation effect of the similar emotion through smiley facial expression, on the social network where users interact each other through web-based user interface rather than in-person interaction. SmileWave is a picture-based networking service and detects the change of smile degree when the user looks at posted smile images of others. Our extensive user study with 50 participants for 30 days confirmed the emotional contagion effect on SmileWave. Users' smile degree improved by 27% when the user looked at posted smile images. The result also proved that there is a stronger effect on smile-based emotional contagion when the examinee and the person in the image are in close relationship.

Poster: NiSleep: Gamification-Friendly Quantitative Evaluation Methodology of Sleep Quality

Kazuki Egashira, Yuki Furukawa, Yuuki Nishiyama, Tadashi Okoshi, Jin Nakazawa, Hideyuki Tokuda

Proceedings of the 14th Annual International Conference on Mobile Systems, Applications, and Services Companion

A robot control system for video streaming services by using dynamic encoded QR codes

Masaki Ogawa, Takuro Yonezawa, Yuuki Nishiyama, Jin Nakazawa, Hideyuki Tokuda

2015 Eighth International Conference on Mobile Computing and Ubiquitous Networking (ICMU)

We propose a novel robot control system by transmitting robot control information on existing video streaming services as dynamic encoded two-dimentional visual code. We implemented sensor data transmitting system by using dynamic encoded two-dimentional visual code which called SENSe-TREAM [1] and we built the robot controlling system by using SENSeTREAM architecture. This paper shows the architecture of robot controlling system and future vision of telepresence and human-robot interaction.

Senbay: Smartphone-Based Activity Capturing and Sharing Using Sensor-Federated Video

Yuuki Nishiyama, Tadashi Okoshi, Yin Chen, Takuro Yonezawa, Jin Nakazawa, Hideyuki Tokuda

Adjunct Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2015 ACM International Symposium on Wearable Computers

We present Senbay, a novel smartphone-based platform for capturing and sharing synchronously-recorded video/sensor data stream using animated two-dimensional barcode. In the ubicomp environment, mobile devices have enabled us to capture video and various types of sensor data easily. However, currently we do not have a framework to instantly integrate video and those sensor data on smartphones and share them easily with other users on the net. In this paper, we present Senbay, a platform to integrate video and sensor data stream into single video on smartphones. With Senbay, sensor data stream will be embedded into a video as animated QR code in real-time, thus users can share the "sensor-federated" video easily on the web, through lots of popular video sharing services. This paper introduces the architecture and evaluation of Senbay and discusses future vision of the sensor-federated video sharing platform along with several applications.

DALT: Detection Algorithm of Throwing Form Changing to Prevent the Baseball Player ’s Throwing Related Injuries

Yuuki Nishiyama, Tomotaka Ito, Takuro Yonezawa, Jin Nakazawa, Kazunori Takashio, Hideyuki Tokuda

UbiHealth: 6th International Workshop on Ubiquitous Health and Wellness (Part of Pervasive 2012 conference)

Towards Measuring Postural Sway Using Earable Devices

Qize Feng, Yanzhi Li, Akihito Taya, Sezaki Kaoru, Yuuki Nishiyama

研究報告ユビキタスコンピューティングシステム（UBI）

ウェアラブルUWBの時系列CIRを用いた呼吸推定

芝塚育大, 藤重凱人, 坪内孝太, 西山勇毅, 下坂正倫

情報処理学会ユビキタスコンピューティングシステム研究会 (IPSJ SIGUBI)

呼吸波形や換気量は健康や認知負荷などを評価する重要な生理指標であるが，測定には一般に装着負荷の大きい機器が必要となる．一方，無線信号の反射から胸郭変動を捉えられるUltra-Wideband（UWB）は，低負荷で呼吸波形まで再現可能な有力な計測手法として注目されている．しかし，従来の研究では，環境設置型での計測を前提としており，ウェアラブルに適用した場合，UWBレーダと体との相対的な位置や向きによる影響や，体動によって呼吸信号が埋もれるなどの課題については十分に議論されていない．本研究では，日常的に装着可能な形態で呼吸波形を計測することを目的として，手首装着型UWBを提案する．この計測形態では，手の配置や姿勢によるレーダと体との向き・距離の違い，および手の動作に伴うセンサ位置の変動が呼吸波形推定に影響を及ぼす可能性があるため，その影響を実験的に検証する．被験者の特定の姿勢・動作下でUWBの時系列CIRから呼吸波形を推定し，環境設置型と手首装着型とを比較して，呼吸波形復元の精度を評価する．

Recognizing Hidden-in-the-Ear Silent Spellings

Xuefu Dong, Yifei Chen, Yuuki Nishiyama, Kaoru Sezaki, Yuntao Wang, Ken Christofferson, Alex Mariakakis

UBIシンポジウム2025

常時装着型イヤラブルデバイス利用環境における音声通知タイミング最適化に向けた基礎検討

伊藤愛香, 小嶋優未, 坪内考太, 西尾信彦, 下坂正倫, 田谷昭仁, 瀬崎薫, 西山勇毅

UBIシンポジウム2025

日常行動センサデータを用いた運動改善のための目標設定支援にむけて

中村仁哉, 田谷昭仁, 瀬崎薫, 西山勇毅

ヒューマンコンピュータインタラクション（HCI）研究会

近年，デスクワークの増加等の理由から，運動不足が社会的な課題になっている．運動不足改善のための目標設定では，達成可能な難易度の短期目標を設定することにより，自己効力感を向上させ，目標達成率の向上に貢献することが明らかになっているが，個人で効果的な短期目標を設定することは容易では無い．また，近年のモバイル・ウェアラブルデバイスの急速な浸透により，個人の行動を容易に計測できるようになったが，運動不足改善のための目標設定において，行動データを用いた目標設定支援技術はまだ十分に確立していない．本研究の目的は，対象者の属性に類似した他人の行動データを短期目標として設定することで，短期目標に対する自己効力感が向上するかを明らかにすることである．本論文では，131 人の一般人から事前に収集した歩数データをもとに，対象者の属性と時刻毎の歩数パターンが類似し，かつ歩数が多い者を抽出し，その者の歩数を短期目標として提示する機構を設計・開発した．10 名の実験参加者に対して，前述の機構を用いて抽出した短期目標を提示したところ，実験参加者は提案に対し高い自己効力感を示した．

Expression Recognition Based on Ear Canal Shape Detection Using Earbud and Ultrasound

高天, 董学甫, 田谷昭仁, 西山勇毅, 瀬崎薫

第24回情報科学技術フォーラム (FIT2025)

常時装着型イヤラブルデバイス利用環境における音声通知タイミング最適化に向けた基礎検討

伊藤愛香, 坪内孝太, 西尾信彦, 下坂正倫, 田谷昭仁, 瀬崎薫, 西山勇毅

研究報告ユビキタスコンピューティングシステム（UBI）

近年, イヤホンやヘッドホンのような耳に常時装着するイヤラブルデバイスが広く普及し, その発展に伴い音声エージェント, 音声ナビゲーションや音声検索など, イヤラブルデバイスを介した音声による情報提供機会が増加している. イヤラブルデバイスの利用シーンとして屋外の歩行中が挙げられるが, 歩行中に通知を受け取ることは周囲への注意力の低下に影響するため, ユーザが受信可能なタイミングで通知を行う必要がある. 先行研究では画面通知の最適化について幅広く研究されているが, 音声通知に着目した研究は相対的に不足している. そこで, 本研究は歩行中という利用状況に焦点を当て, スマートフォンおよびイヤラブルデバイスのセンサデータを活用した音声通知のタイミング最適化について検討する. 歩行中の音声通知受信可否タイミングについての基礎調査の結果, 他の交通との衝突を回避する状況での音声通知は受信拒否されることが分かった. 本稿では, データ取得のために開発したアプリケーションと基礎調査の分析結果, およびモーションセンサデータを用いた音声通知受信可否タイミング検出アルゴリズムの可能性について報告する.

パッシブモバイルセンシングを用いた育児ノイローゼの検知に向けた基礎的調査

顧修聞, 田谷昭仁, 西山勇毅, 瀬崎薫

研究報告ユビキタスコンピューティングシステム（UBI）

育児ノイローゼは、育児の過程で発生する慢性的なストレスによって引き起こされる心理的な状態である． 育児ノイローゼの悪化は産後うつをはじめとする，うつ病に進行する可能性があり，早期発見が重要である． パッシブモバイルセンシングを用いたうつ症状の検知手法は数多く提案されているが，未就学児を育てる父母を対象とした研究は限られており，その行動パターンや心理状態に関する客観的な情報が不足している． そこで本研究では，パッシブモバイルセンシングを用いた育児ノイローゼの検知システムの構築に向けて，未就学児を育てる父母を含む，135名から行動データおよび心理状態を収集し，断面分析を行なった． 特に歩数，位置情報，通話頻度，心理状態の分析を行い，未就学児を育てる家庭の行動パターンの特徴を明らかにした．

学生アスリートの行動データを用いた心身のストレス・回復状態の検知に向けて:異なる時期の比較分析

王振博, 田谷昭仁, 加藤 貴昭, 西山勇毅, 瀬崎薫

情報処理学会 第86回全国大会

パッシブモバイルセンシングを用いた産後うつ症状の検知に関する一検討

顧修聞, 田谷昭仁, 瀬崎薫, 西山勇毅

情報処理学会 第86回全国大会

Initial Investigation on Improving AED Delivery Efficiency by Encouraging Location Awareness

Helinyi Peng, Akihito Taya, Yuuki Nishiyama, Kaoru Sezaki

電子情報通信学会2024年総合大会

GNSS衛星ごとの信号情報に対する点群ニューラルネットワークを用いたUVインデックス推定

厚見昴, 石岡陸, 坪内孝太, 西山勇毅, 瀬崎薫

研究報告ユビキタスコンピューティングシステム（UBI）

個人の曝露した紫外線量の推定手法として，スマートフォンでGNSS衛星から受信した信号情報を用いる方法が研究されている．既存手法では衛星を天球上での位置でグループ化して信号情報をグループ内統計値で代表するため，衛星単位の情報が失われるとともに衛星同士の位置関係の情報も利用できない．そこで，本研究では点群ニューラルネットワークを用いて衛星ごとの信号情報とその近傍関係を直接利用するUVインデックス推定手法を提案する．同一地域の2ヶ所においてGNSS信号とUVインデックスのデータを収集して検証した結果，提案手法が推定精度を向上させると示唆された．衛星ごとの信号情報を活用する手法が発展することで，実世界の多様な環境における高精度な紫外線量推定の実現が期待できる．

学生アスリートの行動データを用いた心身のストレス・回復状態の検知に向けて：行動データと心理尺度との関係分析

王振博, 田谷昭仁, 加藤貴昭, 瀬崎薫, 西山勇毅

研究報告ユビキタスコンピューティングシステム（UBI）

競技と学業の両立を目指す学生アスリートにとって，自らの心身の心理的ストレスと回復状況の精緻 な把握は極めて重要である．既存研究においては，身体活動や心身健康の調査は主に主観的な報告に依存 しており，持続的で客観的なデータのサポートが不足しているため，ある程度の偏りと制限性が存在する． さらに，学生アスリートの特有のニーズやライフスタイルに対する研究は相対的に不足している．そこで 本研究では，スマートフォン・ウェアラブルデバイスに搭載されたセンサを活用して，学生アスリートの競 技中の行動データを途切れることなく継続的に追跡する．多角的なデータ分析を通じて，学生アスリート の行動データと心身のストレス・回復状態との関連性を分析する．また，競技中のアスリートの運動量，各 睡眠ステージの時間，各異なる時間帯のデータと心身のストレス・回復状態との関連性を詳細に分析する．

ハンズフリーのデバイス操作のための汎用イヤラブルデバイスのIMUセンサーを用いた表情認識手法

北森迪耶, 坪内孝太, 西尾信彦, 西山勇毅, 下坂正倫

研究報告ユビキタスコンピューティングシステム（UBI）

本研究では，汎用イヤラブルデバイスに内蔵される IMU センサーを用いて，表情認識が可能となる手法を提案する．表情認識は，ハンズフリーでのデバイス操作や障害者支援システムなど，幅広い応用が期待できる研究分野である．しかし，先行研究では，カメラで撮影した画像を用いた表情認識や外部センサーを取り付けたカスタムデバイスを使用した表情認識が盛んであり，プライバシー問題や日常生活動作の障害となる可能性を孕んでいる．そのため，表情認識技術は商業的な利用が多く，日常生活での活用事例は少ない．そこで，本研究では，表情にコマンドを割り当てることで日常的なハンズフリーデバイス操作を可能とするためのイヤラブルデバイスによる表情認識手法を提案する．提案手法では，汎用イヤラブルデバイスである ”AirPods” を使用して，8 人の被験者から 5 種類の表情変化を記録した時系列データを用いて，機械学習モデルによる表情認識の評価実験を行う．その際，訓練データに表情認識対象者のデータを含める場合と含めない場合の 2 種類の評価実験により，ユーザー依存性の有無も考慮した汎用イヤラブルデバイスでの表情認識の可能性を示す．

ユーザの移動性を活用した地域連携型連合学習における学習モデル統合手法の評価

小野翔多, 山崎託, 三好匠, 田谷昭仁, 西山勇毅, 瀬崎薫

第31回 マルチメディア通信と分散処理ワークショップ (DPSWS2023)

近年，クラウドやエッジなどの計算施設でセンサシングデータを分析することは多様なサービス提供に利用されているが，設備コストや電力消費の増加が問題となっている． これに対処するため，高性能な市販デバイスを活用し，モバイルデバイス同士で学習を協力して実行する連合学習が提案されている． 連合学習は大規模施設の設置費用や消費電力を削減でき，ユーザのプライバシーも保護できる． さらに，地域特有のデータ特性を考慮に入れた地域限定型連合学習も提案されているが，これは特定の地域を一つだけに限定して実行され，地域間の連携は考慮されていない． そこで，本稿では，ユーザの移動性を活用して，地域ごとの学習モデルを統合する地域連携型連合学習を提案する． この手法は，特定の地域で学習されたモデルを他の地域と統合し，それによって学習モデルの性能を向上させることを目指す． 提案手法により，地域ごとに特性のあるモデルを作成しつつ，必要に応じて地域間での学習モデルの統合が可能となり，学習モデルの性能向上が期待できる． 評価の結果，学習モデルの統合は一時的に学習精度が低下するが，追加学習によってモデルの精度が向上することが確認された．

対話中の非言語行動と大規模言語モデルを活用したシームレスな質疑応答補助システムの提案

伊藤愛香, 厚見昴, Wang Zhenbo, Gu Xiuwen, 田谷昭仁, 西山勇毅, 瀬崎薫

第31回 マルチメディア通信と分散処理ワークショップ (DPSWS2023)

領域的情報と説明的情報を用いた生成型画像圧縮伝送に関する初期検討

細沼恵里, 山崎託, 三好匠, 田谷昭仁, 西山勇毅, 瀬崎薫

2023年電子情報通信学会通信ソサイエティ大会

空間内を撮影した画像をネットワークを介して伝送するシステムは多々存在する．しかし，将来的な通信端末数の増加に伴い，今後，多地点で撮影した画像を同時に伝送することが困難になる可能性がある．これに対し，様々な画像圧縮技術が提案されているが，これらの手法は画像内の全情報を維持したまま圧縮を行う．しかし，場面によっては，画像内の全情報を維持して圧縮することが冗長となる可能性がある．そこで，画像から特定の情報を抽出して伝送し，受信端末が画像を再生成することでデータ量を削減できると考えられる．本稿では，画像から特定の情報を抽出して伝送し，受信端末が画像生成モデルを用いて画像を復元する手法を提案する．

連合学習におけるスマートフォンの電力消費量の調査

小野翔多, 山崎託, 三好匠, 田谷昭仁, 西山勇毅, 瀬崎薫

2023年電子情報通信学会通信ソサイエティ大会

集中管理型の連合学習では，学習のための計算量が大規模サーバやデータセンタなどの中央サーバに集中するとともに，データ集約のための通信量が増加する問題がある． そこで，学習クライアントとしてモバイルデバイスを用いるユーザ参加型連合学習が提案されている． 本手法により，前述の計算量と通信量の問題を解決できるが，モバイルデバイスの電力消費量の増加が懸念される．モバイルデバイスは通常，電源に接続されておらずバッテリー駆動であることが多いため，学習に参加する際，電力消費量とバッテリー残量を考慮する必要がある． 本稿では，モバイルデバイスとしてスマートフォンを用い，ユーザ参加型連合学習のアプリケーションを実装して，電力消費量とバッテリー残量が学習性能に与える影響を評価する．

パッシブモバイルセンシングを用いた学生アスリートのコンディション検知に向けた基礎調査

西山勇毅, 加藤貴昭, 瀬崎薫

研究報告ユビキタスコンピューティングシステム（UBI）

学生アスリートにとって心身のストレスとその回復状態を手軽に認識できることは，競技と学業生活を健康に過ごす上で非常に重要である．既存研究では，アンケート調査や血液検査，高性能な生体センサを用いて心身のストレス状態の計測が行われているが，計測負荷が大きく継続利用は難しい．そこで本研究では，市販のスマートフォン・ウェアラブルデバイスに搭載されたセンサを活用し，低負荷にアスリートのコンディションを検出するシステムを開発する．特に本稿では，アスリートのコンディション検知に向けて，データ収集基盤の設計と実装する．さらにデータ収集実験を実施し，収集データからコンディション検知に機構に向けた基礎的な調査を行う．

異なる環境条件におけるGNSS信号強度とUVインデックスの関係

石岡陸, 坪内孝太, 西山勇毅, 瀬崎薫

研究報告ユビキタスコンピューティングシステム（UBI）

近年，個人の浴びる紫外線（UV）量の推定手法として，スマートフォンの衛星測位システム（GNSS）センサを用いる方法が研究され始めた．GNSS による推定は，ユーザの負担なくUV 量を推定できる可能性があるという原理的な利点を持っている一方で，十分な有効性が示されている環境は限定的である．そこで，本研究では，実際のユースケースに近い形でデータを収集できるシステムを考案し，そのシステムを用いて，3 つの地域・2 つの時期・4 つの収集形態（頭の上・ポケットの中・リュックの中・地面に固定）でデータを収集した．さらに，収集したデータを用いて，UV 量推定に向けた基礎的検討として，衛星の信号強度とUV インデックスの相関を計算し，異なる条件下での相関の違いについて比較を行った．分析結果のうち，特に，スマートフォンがリュックの中にあっても相関があること，衛星によって相関の強さが大きく異なることは，今後のGNSS によるUV 量推定の精度改善につながると期待される．

自動運転車と歩行者間の合意形成手法の基礎的検討

牛島秀暢, 石岡陸, 田谷昭仁, 西山勇毅, 瀬崎薫

電子情報通信学会 総合大会

自動運転社会の到来から、これまで運転者と歩行者が暗黙的に行ってきた合意形成が行われなくなる事から、これまで起き得なかったような事故やトラブルが発生すると考えられている。そのため、新しい自動運転車と歩行者間での合意形成手法が必要となる。本研究では、狭路の一方通行における追い越し場面を想定し、音声情報と視覚情報を複合的に組み合わせた自動運転車と歩行者との合意形成手法の基礎的検討をした。

LPWAによる屋内空間の混雑領域推定に向けた検討

細沼恵里, 三好匠, 山崎託, 西山勇毅, 瀬崎薫

2023年電子情報通信学会総合大会

空間内に設置した無線センサノードが送受信する信号の受信信号強度（RSSI: Received Signal Strength Indication）に基づき混雑度を推定する手法が提案されている．これらの手法では，室内やイベント会場内などの開けた領域にセンサノードを設置し，定期的に制御メッセージを交換することで，各ノードが受信したメッセージのRSSIに基づき領域内の滞在人数を推定する． しかし，これらの手法によって，上述する領域が複数存在する建物のフロア全体など，広域な屋内空間内の混雑領域を推定するためには，各領域内に多数のノードを設置する必要がある．そこで，本稿では，低消費電力かつ広域通信が可能な無線通信規格であるLPWA（Low Power Wide Area）を用いた低コストかつ様々な広域屋内環境に適用可能な混雑領域推定手法の実現に向けて，LPWAノードを用いたRSSIの計測実験を行う．

[奨励講演] スポット型連合学習におけるユーザ滞在時間が学習性能に与える影響の評価

小野翔多, 山崎託, 三好匠, 田谷昭仁, 西山勇毅, 瀬崎薫

電子情報通信学会 情報通信マネジメント研究会（ICM）

連合学習では，モデルの更新のたびに通信が発生し，モバイルネットワークの通信量が増加する.そこで， 人が集まりやすい特定のスポットに着目し，そこに集まる人たちのみでデバイス間直接通信を用いて学習するスポッ ト型連合学習を提案する.提案手法は，特定のスポットに適したデータを学習し，周辺ユーザに対してサービスを提 供できるが，移動によってスポットに存在するユーザ数が変動する.本稿では，実装実験によってユーザの滞在時間 が学習に与える影響を評価した.実験の結果，デバイスの離脱が学習性能を低下させる傾向があることが確認できた.

広域屋内空間における人の滞在が受信信号強度に与える影響の解析

細沼恵里, 山崎託, 三好匠, 田谷昭仁, 西山勇毅, 瀬崎薫

電子情報通信学会 情報通信マネジメント研究会（ICM）

快適かつ安全な都市空間を維持するため，空間内の混雑度を推定する様々な手法が提案されている． 特に，導入コストやプライバシの観点から，無線センサノード同士が送受信するメッセージの受信信号強度の変化を基に空間内の滞在人数を推定する手法が提案されている．しかし，従来手法は，一つの部屋やイベント会場など，ある特定の空間内での利用を想定している． そのため，建物のフロア全体などの広域な屋内空間内の混雑度とその領域を推定するためには，各部屋に多数のノードを設置する必要がある．そこで，著者らは，広域通信が可能なLPWA（Low Power Wide Area）を用いることにより，少ないノード台数で混雑領域を推定する手法の実現を目指している．本稿では，混雑領域の推定に利用可能な指標を調査するため，広域屋内空間における人の滞在場所がLPWAノードの受信信号強度に与える影響を実験により解析する．

無線アドホックネットワークにおけるユーザ参加型連合学習の実装実験

小野翔多, 山崎託, 三好匠, 田谷昭仁, 西山勇毅, 瀬崎薫

2023 電子情報通信学会総合大会

ウェアラブルデバイスを用いた乳幼児コンテクストの検知に向けた一検討

小野寺文香, 石岡陸, 西山勇毅, 瀬崎薫

情報処理学会 第85回全国大会（電気通信大学）

Generic Trip Purpose Inference Modelling on Trip Chain

Suxing Lyu, Yuuki Nishiyama, Kaoru Sezaki, Takahiko Kusakabe

研究報告ユビキタスコンピューティングシステム（UBI）

Urban transportation systems are under increasing pressure from rapid population growth. At the same time, urban functions are becoming complex and diversified. In this context, the rapid transformation of cities has long promoted the demand for human mobility (travel behavior) analysis. Trip purpose, one of the behavioral factors, is crucial to understanding human mobility generation. But keeping track of trip purpose is not easy. Nowadays, there is a huge volume of human mobility data passively collected by mobile devices. Trip purpose is right the missing item form the data. The more reliable we are in the inference of the missing items, the more beneficial we can get from the data. A generic trip purpose inference can bring semantic information to human mobility. Consequently, the decision of urban transportation and urban infrastructure development will be improved and supported based on the understanding of human mobility generation. The doctoral study reports the findings on developing the generic privacy-insensitive trip purpose inference for one-day travel (trip chain).

スマートフォンのGNSSセンサを用いたUVインデックス推定

石岡陸, 坪内孝太, 西山勇毅, 瀬崎薫

研究報告ユビキタスコンピューティングシステム（UBI）

個人の紫外線被曝量モニタリングは，個々人が最適な量の紫外線を浴びることを目的に，盛んに研究されてきた．先行研究では，スマートフォンのカメラや光センサ，ウェアラブルUVセンサを用いた UV 量の推定が提案されてきた．我々は，スマートフォンの GNSS センサを活用した世界初の UV インデックス推定手法を提案する．カメラや光センサを継続的に露出させなければならないような手法と比較して，GNSS を用いた方法は，普段のようにスマートフォンを身につけるだけで推定を可能にするという原理的な利点がある．GNSS による推定の第一歩として，1 つの地域の 3 ヶ所において GNSS データを収集し，OpenUV というAPI をベースラインとして用い，提案手法の有効性を検証した．提案手法は，平均絶対誤差（MAE）0.1523 を達成し，ベースラインを圧倒的に上回った．本研究によって実際のデータに対して提案システムの有効性が証明されたことは，人々が容易に日々の UV 被曝量を把握できるような未来への大きな一歩となることが期待される．

A preliminary study for monitoring hygiene behaviors by using multiple sensors on a wrist

Haoyu Zhuang, Liqiang Xu, Yuuki Nishiyama, Kaoru Sezaki

研究報告ユビキタスコンピューティングシステム（UBI）

Under the epidemic of COVID-19, it is important to automatically detect epidemic protective behaviors without a user's intention. Existing studies utilized only sensor data from IMU for detecting epidemic protection behaviors. However, the performance of the classification for similar behaviors could be unsatisfactory due to the single data dimension. It is well known that washing hands and hand sterilization are essential personal hygiene behaviors. In this paper, we use multiple sensor data from an off-the-shelf smartwatch and smartphone for detecting these three behaviors. Our performance evaluation indicated that our proposed method has improved accuracy for classifying the target epidemic protective behaviors over previous methods. Furthermore, for applying our method in reality, we developed a prototype for detecting these behaviors on a wearable device, which allows us to utilize our method widely in health habits monitoring.

Using Turning Data for Micro-Mobility Rider Identification

韓増易, 西山勇毅, 瀬崎薫

電子情報通信学会ソサイエティ大会

Face-Touch Detection with Smartwatch by CNN: An Experimental Lab Study

荘昊昱, 韓増易, 西山勇毅, 瀬崎薫

電子情報通信学会ソサイエティ大会

LPWAによる屋内混雑度推定に向けた基礎検討

細沼恵里, 三好匠, 山崎託, 西山勇毅, 瀬崎薫

電子情報通信学会ソサイエティ大会

無線マルチホップ連合学習へ向けた実装実験

小野翔多, 三好匠, 山崎託, 西山勇毅, 瀬崎薫

電子情報通信学会ソサイエティ大会

イアラブルデバイスを用いた街歩き時におけるユーザの道迷い状態の検知

下条和暉, 西山勇毅, 瀬崎薫

研究報告ユビキタスコンピューティングシステム（UBI）

街歩きにおいて，ユーザの迷い状態を検知できれば効率的で安全な道案内が実現出来る．イアラブルデバイスは，耳に装着するウェアラブルデバイスであり，搭載された行動認知機構や音声インターフェスによってユーザの状態を把握することが出来る．イアラブルデバイスを活用することえユーザの街中における道迷い状態を検知し，音声案内が可能になると考えられるが，その手法についてはまだ明らかになっていない．そこで本研究では，イアラブルデバイスに搭載されているモーションセンサから収集したデータを用いることで，ユーザの道迷い状態を検知する手法を提案する．

スマートウォッチを用いたマスク装着の促進手法

小野翔多, 西山勇毅, 瀬崎薫

情報処理学会 IoT行動変容学研究グループ キックオフシンポジウム

新型コロナウイルス感染症が世界的に蔓延しており，感染拡大を予防するために，マスクの装着などの飛沫感染のリスクを下げる行動が求められている [1]．感染リスクを低下させるためには、常時マスクを装着することが望ましいが，無意識のうちにマスク非装着のまま行動してしまうことがしばしば発生する．マスク装着を効果的に促すためには，個々人のマスク装着状態を自動検知し，その状態に応じてユーザに行動変容を促すことが求められる．しかし，これまでのところ，マスク装着状態を市販の端末のみを用いて常時検出する手法はまだ提案されていない．本稿では，スマートウォッチに搭載されている複数のセンサを用いてマスク装着状態を自動検知し，マスク非装着のユーザに通知することで，マスク装着の行動を促進させる手法を提案する．

タクシー車両を用いたマイクロモビリティ再配置

牛島秀暢, 西山勇毅, 瀬崎薫

情報処理学会 第84回全国大会（愛媛大学）

腕時計型ウェアラブルデバイスを用いた会話時間計測手法の構築に向けて

小松勇輝, 下条和暉, 西山勇毅, 瀬崎薫

情報処理学会 第84回全国大会（愛媛大学）

スマートフォンを用いたベビーカーのコンテキスト検知に向けた一検討

西山勇毅, 瀬崎薫

電子情報通信学会総合大会（オンライン）

ベビーカーは，子供がいる家庭の約70%が所有し，週に一回以上利用する家庭は73.8%であると報告されるなど，保有率・利用率ともに高い．ベビーカーは乳幼児を運搬するため，安全性と快適性が高く求められるが，街中・施設内において安全・快適な走行・滞在可能なルートや場所，時間を手軽に知ることは難しい．これらの情報を検知することで，様々な応用アプリケーションを実現できる．そこで本稿では，スマートフォンを用いたベビーカー移動時の急停止や段差との衝突，路面状態といったベビーカー移動に関するコンテキスト検知の可能性を調査し，その結果を報告する．

ウェアラブルデバイスのマイクを用いたマスク装着状態の検知に向けて

小野翔多, 西山勇毅, 瀬崎薫

研究報告ユビキタスコンピューティングシステム（UBI）

感染症予防において，マスクの装着は飛沫による感染症への感染リスクを低下させる有効な手段の一つである．日常生活中におけるマスク装着の有無やその種類を自動的に検出できれば，感染リスクの判定やJust-in-Timeでの注意喚起，行動記録など様々な応用サービスが実現可能になる．しかし，映像処理や専用機器を用いずに，日常生活中において自動的にマスクの装着状態を検知する手法はまだ提案されていない．本研究では，市販のスマートウォッチの内蔵マイクのみを用いてマスクの装着状態を検出する．マスク装着時の音声特性調査とマスク装着状態判定モデルの評価実験から，マスク装着時・未装着時の音声データと機械学習を用いてマスク装着状態を検知できる可能性が示唆された．

移動体通信併用型MANETにおける端末密度を用いた中継領域制御

小野翔多, 山崎託, 三好匠, 西山勇毅, 瀬崎薫

情報通信マネジメント研究会 (ICM)

Mobile ad-hoc network（MANET）は，端末間の通信のみで自律分散的にネットワークを構築できる．しかし，通信ネットワーク構築時に制御メッセージを全方位にフラッディングするため，通信資源を過剰に消費する. 位置情報を収集できる端末が通信ネットワーク構築に参加する環境では，端末の位置情報を組み合わせることで，より効率的なフラッディングが可能になると考えられる．本稿では，端末の位置情報に基づいて仮想通信領域を作成し，その領域内の端末にのみフラッディングすることで，経路探索を効率化する手法を提案する．シミュレーションによる評価の結果，提案手法は仮想通信領域を柔軟に作成し，過剰な通信資源消費の抑制と通信遅延の削減を実現できることが分かった．

ウェアラブルデバイスを用いた子育てコンテキストの検知に向けて

笠原有貴, 西山勇毅, 瀬崎薫

研究報告ヒューマンコンピュータインタラクション研究会（CHI）

女性の社会進出や核家族化，産後うつ問題など，子育て環境は大きく変化しており，子育ての効率 化や子育て支援は社会的に大きな課題となっている．本研究では，近年普及傾向にあるウェアラブルデバイスを用いて，ミルクやオムツ替え，お散歩など「親」が「乳幼児」に行う子育て行動の検知技術の開発を行う．子育て中のモーションデータを腕時計型のウェアラブルデバイスに搭載されたモーションセンサを用いて収集し，収集データと機械学習を用いて子育てコンテキストの検知モデルを構築する．本稿では，9種類の子育てコンテキストを定義し，子育てコンテキストの検知モデルの構築とその精度評価を行なった．

イヤラブルデバイスを用いた身体感覚記録・利活用システムの構築に向けて

西山勇毅, 瀬崎薫

研究報告ユビキタスコンピューティングシステム（UBI）

スポーツや楽器の演奏，自動車の運転など，効率的に新しい運動スキルを習得し，さらに向上させることは，人々の生活をより豊かにする．運動学習は，主観的な運動感覚と実際の動作とのズレを反復練習により埋める作業であるが，客観的な運動情報に比べ，主観的な情報を低負荷に記録し，それらを活用する環境は整っていない．そこで本研究では，運動学習時における主観的な運動情報を容易に収集・利活用可能なシステムを設計・実装し，評価した．

An Online Task Offloading Strategy in Vehicular Edge Computing

Hong Duc Nguyen, Shunsuke Aoki, Yuuki Nishiyama, Kaoru Sezaki

IEICE Society Conference 2021

移動体通信併用形MANETにおける経路要求送信回数に基づく経路構築手法

小野翔多, 山崎託, 三好匠, 西山勇毅, 瀬崎薫

2021 電子情報通信学会ソサイエティ大会

スマートウォッチを用いた子育て行動の推定に向けた一検討

笠原有貴, 西山勇毅, 瀬崎薫

計測自動制御学会 計測部門スマートセンシングシステム部会

常時装着型イアラブルデバイスにおける割り込み可能タイミングの検討

下条和暉, 西山勇毅, 瀬崎薫

CSIS DAYS 2021

集約型都市におけるライドシェアサービス導入効果のシミュレーション

小池優太郎, 西山勇毅, 瀬崎薫

CSIS DAYS 2021

部屋毎の滞在時間特性を考慮した頑健な滞在場所推定手法

鈴木凌斗, 村上弘晃, 西山勇毅, 川原圭博, 瀬崎薫

研究報告ユビキタスコンピューティングシステム（UBI）

屋内での滞在情報を正確に把握することで，混雑度の推定や集客情報，人流の把握など，様々なサービスを提供できる．BluetoothビーコンやWiFiの信号強度を用いた滞在場所推定手法では，低コストに滞在推定システムを導入できる．しかしながら，受信信号強度の不安定さや隣接した部屋から漏れる信号などが原因となり，単純な信号強度のみを用いた判定では，受信環境によっては滞在場所の誤判定が頻繁に発生する．本稿では，部屋ごとの滞在時間特性の違いを考慮に入れることにより誤判定を抑制する手法を提案する．提案手法では，部屋ごとの滞在時間の分布をワイブル分布にフィッティングし，生存時間解析を適用することによりユーザの状態を推定する．信号強度の強弱のみに基づく既存手法との比較のため，正解ラベル付きのデータを収集し評価実験を行った．

感染症予防行動を促進させるインセンティブモデルに関する一検討

陳美怡, 幡井皓介, 西山勇毅, 瀬崎薫

研究報告ユビキタスコンピューティングシステム（UBI）

現在，新型コロナウイルス感染症（COVID-19）の感染が拡大しており，人々の生命と健康を大きく脅かしている．政府は地方自治体，保健機関は，「手洗い」や「マスクの着用」 「行動記録」「外出自粛」などの感染症予防策を人々に積極的に取り続けることを推奨している.本研究では，ユーザの感染症予防行動の促進を日標とし，既存の行動記録アプリ（SelfGuard）を拡張し，感染症予防行動に対する最適なインセンティブモデルの導入を検討する.具体的には，スマートフォンとウェアラブルデバイスに搭載されたセンサを利用してユーザの感染症予防行動を認識し，行動に応じてインセンティブとして換金可能なポイントを付与する.固定・加算・減算モデルという三種類のインセンティブモデルにおいて人の行動に与える影響の違いを調査する.

感染症予防行動を促進させるインセンティブモデルの構築に向けて

陳美怡, 幡井皓介, 西山勇毅, 瀬崎薫

第20回情報科学技術フォーラム（FIT2021）

現在、新型コロナウイルス感染症（COVID-19）の感染が拡大しており，人々の生命と健康を大きく脅かしている．2021年4月18日時点で，COVID-19による全世界の累計死亡者数が300万人を超えたことが報告された[1]．市や政府は，感染症の感染拡大を防ぐために，自己隔離，ロックダウン，行動制限などの対策を実施している．また政府や保健機関は，人々が手洗い・マスクの着用・行動記録・外出自粛などの感染症予防策を積極的に取り続けることを推奨している．西山らの研究では，行動記録アプリ（SelfGuard）を開発し，半自動的にユーザの滞在情報・行動履歴を記録することで，感染症予防行動の促進を実現している[2]．本研究では，ユーザの感染症予防行動の促進を目標として，既存アプリ（SelfGuard）を拡張し，感染症予防行動に対する最適なインセンティブモデルの導入を検討する．具体的には，スマートフォンとウェアラブルデバイスに搭載されたセンサを利用してユーザの感染症予防行動を認識し，行動に応じてインセンティブとして換金可能なポイントを付与する．定額・加算・減算モデルという三種類のインセンティブモデルにおいて人の行動に与える影響の違いを評価する

BLEビーコンを用いた屋内位置推定システムの設計と実装

山下陸, 西山勇毅, 小松寛弥, 川原圭博

研究報告ユビキタスコンピューティングシステム（UBI）

現在，我々は大学キャンパスでの COVID-19 感染対策，そして，ポストコロナ時代における望ましい位置情報インフラの構築を目指している．ユーザーのプライバシーに配慮し，講義で活用する教室や図書館等の 181 のスペースに 200 の Bluetooth ビーコンを設置した．ビーコンが発する ID をスマートフォンで検知し，ユーザの部屋への入退室を判定する．ユーザの位置情報はサーバーサイドに集約された後，ユーザの同意に基づいたルールで，ニーズに応じた位置情報の開示が可能になっている．設置したビーコンをもとにキャンパス内で位置情報の取得予備実験を行ったところ，室内を隔てる壁の材質によっては電波の減衰率が大きく異なることや，大学の施設管理データベースの登録情報が実態を反映していないことなどが明るみになった．本稿では，MOCHA と名付けた本システムの設計と実装，システムアーキテクチャや利用規約を制定する際のポイントについて共有する． We are currently working to build the desirable location infrastructure on university campus to combat COVID-19 and also in the post-corona era. In consideration of the user privacy, we have installed 200 Bluetooth beacons in 181 spaces, including the libraries and the classrooms used for the lectures. The IDs emitted by the beacons are detected by a smartphone to determine the user's entering and exiting the room. After the user's location information is collected on the server side, rules based on the user's consent enable the location information to be disclosed according to their needs. A preliminary experiment of the location information acquisition in the campus using the beacons revealed that the attenuation rate of radio waves differs greatly depending on the material of the wall, and the information registered in the facility management database of the university does not reflect the reality. In this paper, we share the design and implementation of our system, named MOCHA, and the key points for establishing the system architecture and terms of use.

環境センサを用いたタクシー車室内における感染症リスク評価に関する一検討

牛島秀暢, 西山勇毅, 小野翔多, 瀬崎薫

研究報告モバイルコンピューティングと新社会システム研究会（MBL）

COVID-19の感染拡大により，公共空間や飲食業などの対面サービスにおいて，持続的かつ安全で低コストな感染症への対策が求められている．これまでは，密閉，密集，密接の三密と呼ばれる状態が感染リスクがあるとされ，換気の徹底などが推奨されてきたが，その効果を定量的に計測する事は極めて困難である．本研究では，タクシーという密閉空間の状態を複数の環境センサを用いて計測し，感染症リスクを定量的に分析する．

Non-Negative Tensor Factrization を用いたドックレス型マイクロモビリティの利用形態分類手法の検討

牛島秀暢, 青木俊介, 西山勇毅, 瀬崎薫

研究報告高度交通システムとスマートコミュニティ（ITS）

交通やインフラ，スマートフォンなどから得られる様々なデータを統合的に利活用し，都市計画の継続的な改善に役立てるという都市コンピューティングが注目されている．都市コンピューティングは少子高齢化と過疎化が進行する日本においても公共インフラを有効活用し都市を維持するためにも有効である．限られた公共インフラを活用するためには人々の移動目的を推定し，交通リソースを最適化する必要があるが，既存の IC カードなどの交通データでは推定粒度に限界があった．こうした状況の中，特定の返却場所を持たないドックレス型のマイクロモビリティが急速に普及している．ドックレス型マイクロモビリティは平均移動距離が 500m 程度と短く，直接目的地に向かうため，より詳細な移動行動が検出可能である．本研究では，マイクロモビリティが都市空間で離散的に分布する点に着目した．そして，細かく単発的な移動行動を大域的に分析することで潜在的な移動パターンがあることを，Non-Negative Tensor Factrization と呼ばれる教師なし学習を用いることで明らかにした．

GPS 信号受信状態を用いた紫外線量推定手法の検討

日隈壮一郎, 西山勇毅, 瀬崎薫

研究報告ユビキタスコンピューティングシステム (UBI)

近年の都市構造の変容と急激なライフスタイルの変化に伴い，人間が屋内で過ごす時間は長時間化し，逆に屋外で直射日光を浴びる時間は年々短くなっている．過度な紫外線の被曝は皮膚癌やシワ，シミの発生可能性を高めるが，一方で適度な紫外線被曝は体内でのビタミン D の生成に必要不可欠である．また，基本的に野菜に含まれていないビタミン D の不足は，カルシウム不足や低カルシウム血症，骨の軟化やうつ病などに繋がる危険性があり，長期的な健康管理において，紫外線被曝量の管理は重要である．しかしながら，紫外線センサを常に携帯することはユーザの負担が大きく，長期的な利用には日常的に計測または推定可能な手法が必要である．そこで本研究では，スマートフォンに搭載された GPS モジュールを用いて，GPS 信号の受信状態から紫外線量を推定する手法の検討を行う．

スマートフォンセンサを用いた気分情報を基にした都市空間評価手法に向けて

神村潤, 西山勇毅, 瀬崎薫

電子情報通信学会 電子情報通信学会総合大会

In-the-Wild実験におけるiOS用モバイルセンシングフレームワークの性能評価

西山勇毅, 佐々木航, 栄元優作, 本木悠介, 大越匡, 中澤仁, 瀬崎薫

電子情報通信学会技術研究報告: 信学技報

ユーザの携帯端末より収集したセンサデータを用いて，人や集団・空間の状態を理解するモバイルセンシングは，情報科学だけでなく社会科学や公衆衛生など様々な分野で利用されている．モバイルセンシングを容易に実現するツールは多数提案されており，それらは研究室内実験など「コントール環境」での性能評価は行われている．しかし，実際の被験者にシステムを配布し実験を行う「In-the-Wild環境」における性能評価は行われていない．特にAndroidに比べiOSは制約が多く，制約を無視した設定はデータ収集率の低下を招くため，適切な設定を明らかにする必要がある． そこで本研究では，iOS用モバイルセンシングフレームワーク（it AWARE-iOS）を用いて，1週間のIn-the-Wild環境実験を10人の被験者を対象に行い，その際のit AWARE-iOSのデータ収集性能及びバッテリ消費を評価した．その結果，最も収集率が高い可能性のある設定（ESM＋SPN）では，ユーザがit AWARE-iOSを強制終了しない限り95%以上のデータを収集可能であり，平均11.39時間の起動が可能であることが明らかになった．

Mobile Sensing of Pedestrian Mobility and its Assessment

Chenwei Song, Masaki Ito, Yuuki Nishiyama, Kaoru Sezaki

IEICE Tech. Rep.

We propose a client-server system that provides crowd detection and mobility information. Our proposed system has the advantages of low cost and location flexibility without pre-deployed, as long as there is a sufficient number of users involved. We conducted several experiments in real environments to determine the feasibility, accuracy and applicable environment of the system. The result shows that the system can effectively capture the flow of people in the experimental area. In some cases, under the same environment, it can obtain almost the same mobility tracking information from fewer participating users than the GPS method.

モバイルコンピューティングによるエモーショナル・イーティングの検知

栄元優作, 佐々木航, 西山勇毅, 大越匡, 中澤仁

研究報告ユビキタスコンピューティングシステム (UBI)

ストレスや感情の変化に左右された食行動はエモーショナル・イーティングと呼ばれる．エモーショナル・イーティングが発生する原因として，ストレスや生活習慣の乱れによる自律神経やホルモンバランスの乱れがあげられる．したがって，自律神経の活性度やホルモン分泌量を継続して観察・分析することで，エモーショナル・イーティングの発生を予期することが可能である．しかし，それらの計測には特殊な生体センサが必要であり，また長時間の使用は負担が大きい．本研究では，ユーザが日常的に利用しており，かつ利用者の行動パターンの特徴が表れるスマートフォンのセンサ情報と機械学習を組み合わせることで，エモーショナルイーティングの検知に取り組む．そして，スマートフォンのセンサデータと食生活データを収集する SEED システムを構築し，収集データからエモーショナル・イーティングを検知する機械学習モデルを作成した．本システムを用いて 60 人の被験者から 28 日間のデータを収集したところ，87.5% の精度でエモーショナル・イーティングを検知することに成功した．

人の限定合理性を超越した行動変容支援に向けた情報プラッ トフォームの設計

大越匡, 西山勇毅, 佐々木航, 栄元優作, 中澤仁

研究報告ユビキタスコンピューティングシステム（UBI）

ユビキタスコンピューティングシステムとして，例えば睡眠，食事，運動，学びといった人の多くの行動における行動変容を支援するための研究が数多くなされてきた一方，現状の取り組みは，人間が合理的であろうとするが種々の能力の限界によって限られた合理性しかもてない「限定合理性」(bounded rationality）を超えて人間を支援できていない．本研究では，この限定合理性を超越する情報技術の実現を目指し，その概念を整理し，解決手法における方針と情報アーキテクチャへの要件を明らかにする．

Likeyboard: ライフログデータ共有簡易化のためのスマートフォンキーボードインタフェース

石川伶, 佐々木航, 西山勇毅, 大越匡, 米澤拓郎, 中澤仁, 高汐一紀, 徳田英幸

研究報告ユビキタスコンピューティングシステム（UBI）

近年，食事や睡眠，運動量などを記録するライフログに対する関心が高まってきている．スマートフォンやウェアラブルデバイスなど，様々なライフログデータを検知可能なデバイスの普及し，様々なライフログデータが蓄積されている．しかし，蓄積されたライフログデータは，ライフログアプリケーションやウェブ API を通して個人の活動の振り返りには利用されているが，共有を行うユーザは少ない．ライフログデータの共有は更なるライフログデータ蓄積のモチベーションを向上や，新たなコミュケーションの創出，行動変容の促進にも繋がる．そこで本研究では，ライフログデータ共有の簡易化を目的とした，“Likeyboard” を提案する．“Likeyboard” は，スマートフォンのキーボードインタフェースを用いて 「ライフログデータの選択」 「日時の選択」 「出力形式の選択」 のみで，スマートフォン上でのライフログデータの共有を可能とする．大学生 12 人を被験者として，既存のライフログデータアプリとのインタフェースの比較実験を行った．その結果本システムが既存アプリよりライフログデータの共有が簡易化されたいう評価結果を得た．

TalkingNemoPad : Aquarium Fish Talks Its Mind for Breeding Support through Your iPad

礒川直大, 西山勇毅, 大越匡, 中澤仁, 高汐一紀, 徳田英幸

研究報告ユビキタスコンピューティングシステム（UBI）

ペットとしての生き物を飼育する際には，飼い主がペットの状態を正しく理解できず，ペットの体調管理は難しくなっている．世界中で最も飼育されているペットである観賞魚においては特に飼い主とペットとのコミュニケーションがより困難であるため，犬や猫などと比較してもそのような状況に陥りやすい．そのため，飼い主と観賞魚とのインタラクションを促進し，観賞魚飼育に役立てる研究が行われているが，それらのシステムは特定の水槽においてのみ動作しているため，一般的な飼育者が用いることは難しく，また様々な種類の観賞魚に関するデータを一律に収集することができない．本稿では，広く普及したデバイス上で動作する飼い主と観賞魚のインタラクションシステム “TalkingNemoPad” を提案する．TalkingNemoPad には，ユーザのタブレット上で動作し，観賞魚を飼育する人の多くが導入 ・ 実用可能という特徴がある．また，これにより様々な観賞魚飼育環境から広くデータを収集することが可能となる．本稿では，TalkingNemoPad が一般的な観賞魚飼育環境下でに使用可能であることを示すために，環境の異なる複数の水槽を用い，本システムの適用可能な観賞魚の飼育環境について評価する．

SpoTrip : 観光客リピータ化促進のための隠れスポット情報提供システムの評価

片山晋, 礒川直大, 小渕幹夫, 西山勇毅, 大越匡, 米澤拓郎, 中澤仁, 高汐一紀, 徳田英幸

電子情報通信学会技術研究報告 = IEICE technical report : 信学技報

HealthFight: 多数決を用いた食習慣改善ソーシャルメディア

栄元優作, 江頭和輝, 河野慎, 西山勇毅, 大越匡, 米澤拓郎, 高汐一紀, 中澤仁

研究報告ユビキタスコンピューティングシステム（UBI）

近年，食習慣の悪化による肥満が社会問題となり，継続的な食習慣の改善手法が求められている．継続して日々の習慣を変えるには動機の維持向上が重要であり，ユーザの同期を持続させる手法として，ゲームのメカニズムを用いたゲーミフィケーション手法が提案されている．ゲーミフィケーション手法の 1 つである 「競争手法」 は，歩数などのユーザの行動量を可視化し，他者との優劣を意識させることで，動機の維持向上を促す．しかし食事分野では，食事指標として知られるカロリー値の正確な算出が難しく，計算には労力がかかるため，競争の適用が困難である．本研究では本問題を解決するために，ユーザ間で食事の健康度合いを競わせるシステム 「HealthFight」 を構築し有効性を評価する．本システムは食事画像を共有するソーシャルメディアであり，投稿した食事画像の 「ヘルシーさ」 をユーザ間で競わせる．競争相手は食事ごとに更新されるため，ユーザは食事ごとに違った楽しさを感じられ，継続した利用が見込める．本システムによる食習慣の変化を評価するため，7 人の被験者に対して 20 日間の実験を行った．その結果，他者の評価を気にしやすい被験者の食生活は，健康的になる傾向が見られた．

NiSleep: ゲーミフィケーションを適用可能な睡眠評価

江頭和輝, 古川侑紀, 西山勇毅, 大越匡, 米澤拓郎, 中澤仁, 高汐一紀, 徳田英幸

研究報告ユビキタスコンピューティングシステム（UBI）

人気ゲームでよく使われるゲームデザインの技術を現実の問題解決に用いることをゲーミフィケーションと呼ぶ．しかし睡眠に関しては，必ずしも睡眠時間が長いほど良い睡眠であるとは限らない特徴を持つため，既存のゲーミフィケーションを適用することが難しい．本研究ではゲーミフィケーションを適用可能な睡眠評価方式 「NiSleep」 を提案する．社会的な活動時間帯と生物時計の不一致によって生ずる不調を社会的ジェットラグと呼ぶ．そして，NiSleep 指数は社会的ジェットラグが解消される睡眠であるかを 「睡眠時間」 「睡眠中央時刻」 「睡眠効率」 を 100 点法で評価し，それらから最終的な評価点を算出する．NiSleep 指数を用いた睡眠評価を行う NiSleep アプリケーションを構築し，被験者 10 人に対して 28 日間にわたるユーザ環境上での実証評価実験を行った．睡眠評価指数の精度と使用者の睡眠行動の変化について検証した．その結果，提案した睡眠評価指数は社会的ジェットラグの特徴を再現 (r=-.440) した．

MyTime : タイムマネジメントの促進を目的としたアプリケーション使用時間自動記録システムの提案

小出粋玄, 西山勇毅, 大越匡, 米澤拓郎, 高汐一紀, 中澤仁, 徳田英幸

電子情報通信学会技術研究報告 = IEICE technical report : 信学技報

MyFactor : ライフログを用いたユーザの内面状態に関する因子分析

小渕幹夫, 古川侑紀, 西山勇毅, 大越匡, 米澤拓郎, 高汐一紀, 中澤仁, 徳田英幸

電子情報通信学会技術研究報告

Senbay : 活動促進のためのスマートフォンを利用したセンサデータ統合型動画記録・共有・分析プラットフォーム

西山勇毅, 米澤拓郎, 中澤仁, 徳田英幸

知的環境とセンサネットワーク研究会(ASN), 電子情報通信学会技術研究報告 = IEICE technical report : 信学技報

SmileSpot: 他人の笑顔画像の共有によるユーザの笑顔形成への影響評価

佐々木航, 西山勇毅, 大越匡, 米澤拓郎, 中澤仁, 高汐一紀, 徳田英幸

情報処理学会研究報告. UBI, [ユビキタスコンピューティングシステム]

「人は幸福であるが故に笑うのではなく,笑うが故に幸福である」 という言説にもある通り笑顔を形成することがユーザの幸福感につながると考えられ,これを支持する研究は多い.幸福感は周囲に伝達することが知られている.以上のことから,幸福な人からの笑顔情報によるユーザの笑顔形成への促進が,ユーザの幸福感を引き起こし,それによって幸福感が伝達していくと考えられる.そこで本研究では,他人の笑顔画像の共有によるユーザの笑顔形成への影響を評価する "SmileSpot" を実装した."SmileSpot" は個人のスマートフォン端末のディスプレイに他人の顔画像を映し出し,それを見た際のユーザの笑顔度を検知する.38 人の被験者を対象とした 15 日間の評価実験から他人の笑顔画像がユーザの笑顔形成を促進させることがわかった.それに加え,女性のほうが男性と比べて影響を受けやすく,また親密度が高い人の笑顔画像のほうが親密度の低い人の笑顔画像より影響を受けることがわかった. As there is a famous statement, "We don't laugh because we're happy, we're happy because we laugh", we are considered that making users smile will make them happy. There are many research that support this idea. And also, "happiness" is known to transfer to surroundings. Therefore we thought that formating users'smile by smile of others who are happy will make users happy, and this process will transfer happiness to surroundings. Now, in this research, we create a system called "SmileSpot" to see the effect of smile which makes users also smile and lead them to happiness. "SmileSpot" works by showing pictures of people smiling on user's smartphone display, and observe how much users will laugh when they using this system. By an experiment which was done by observing 38 research participants over 15 days, result which proved the statement came up. "Others' smile will make users happy" is proved. In addition, we found that women are more susceptible to influence than men, and also smile-image of the community that users belong to is easier to effect the users than the smile-image which was taken from the community that users don't belong or care.

Aqua Mapping: 水槽を介した観賞魚とのインタラクションシステム

礒川直大, 西山勇毅, 大越匡, 米澤拓郎, 中澤仁, 高汐一紀, 徳田英幸

研究報告ユビキタスコンピューティングシステム（UBI）

魚を飼い，その水槽を鑑賞することは，我々の生活に潤いと落ち着きを与えてくれる．観賞魚を飼育する上で水槽内の環境や魚の体調管理が重要となるが，観賞魚は自分の状態を飼い主に伝えることができないため，飼い主が魚の状態に基づきインタラクティブに飼育することは難しい．本研究では，観賞魚の位置情報をもとに水槽背面のディスプレイに観賞魚の状態を表示することで，飼い主に水槽内の環境や観賞魚の体調などの情報を伝えるシステム “Aqua Mapping” を提案する．本システムにより水槽を介した観賞魚とユーザのインタラクティブな飼育環境を提供する． It relaxes us and enriches our daily lives to rear fish and appreciate an aquarium. In that case, the aquarium environment and the physical condition management of the fish become important. However, it's difficult that an owner rears aquarium fish interactively based on the state of the fish because fish can't tell the owner about its state. Thus, this study proposes the “Aqua Mapping” which informs an owner about the condition of an aquarium and fish by indicating fish state on the screen set behind the aquarium using the position information. This system can offer interactive rearing environment between an user and fish through an aquarium.

動画二次元コードを用いたウェアラブルセンシングのための情報記録・配信手法

米澤拓郎, 西山勇毅, 小川正幹, 中澤仁, 徳田英幸

第19回人間情報学会

加速度センサのみを用いた移動手段判定アルゴリズムの評価

古川侑紀, 西山勇毅, 大越匡, 中澤仁, 高汐一紀, 徳田英幸

情報処理学会研究報告. MBL, [モバイルコンピューティングとユビキタス通信研究会研究報告]

近年のスマートフォンの普及により,様々なセンサを利用できるようになった.それに伴い,日常的な行動を記録する研究が盛んになっている.移動手段の判定もその一部である.既存の研究では加速度センサやマイク,GPS など複数のセンサを用いて移動手段を判定しているため,電池の消耗が激しくスマートフォンのみを用いて判定することは実用的ではない.本研究では加速度センサを用いて移動手段の判定を行うシステムの提案を行い,本システムに様々な機械学習の手法を適用することによって分類精度の評価を行う.また,本システムの省電力性を評価する. Nowadays, various kinds of sensors are available due to expansion of the smartphones. Along with this, the research that records the action on a daily basis has become very popular. Distinction of transport is also a part of the studies. The major result from resent studies has been detected those transport by using the sensors such as GPS or a microphone or an acceleration sensor. However, since this is one of the biggest cause to encourage the consumption of the batteries, it is not practical to detect only with a smartphone. I suggested in this paper the system to detect transport adopting acceleration sensor and evaluated the detection accuracy by applying various kinds of method of machine learning. Moreover, I evaluated power saving of this system.

チームの動機づけにおける個人の貢献度の可視化に関する一検討

西山勇毅, 米澤拓郎, 中澤仁, 徳田英幸

情報処理学会研究報告. UBI, [ユビキタスコンピューティングシステム]

近年,個人の動機付けや行動改善のためのアプリケーションが普及してきた.今後,研究プロジェクトチームやスポーツチームなどの明確な目標を持ち,様々な立場の人間が所属するチームでの利用が考えられる.本研究では,個人の貢献度の可視化がチームの動機づけに与える影響の調査を行った.情報共有アプリケーションを実装し,ランダムに抽出したチームに対して,他人に個人の貢献度を見せた場合と見せない場合での比較実験を行い評価した.

野球選手の投球動作変化を用いた疲労度取得手法の提案 : 投球障害予防支援システム構築に向けて

西山勇毅, 瀧本拓哉, 米澤拓郎, 中澤仁, 高汐一紀, 徳田英幸

電子情報通信学会技術研究報告. USN, ユビキタス・センサネットワーク : IEICE technical report

野球では崩れた投球フォームでの投球や過度の練習などにより,投球障害(肩や肘への怪我)が多く発生する.特に野手は自分の練習量を意識せずに練習することが多く,主観による体調管理しか行われていない.既存研究では,スポーツ障害の防止,投球パフォーマンスの向上などを目的として,三次元モーションキャップチャーや動画解析などによる投球動作の分析が行われてきた.しかし,これらの研究は実際の練習での使用を想定していないため,場所の制限や解析の時間的コストなどの問題点がある.本研究では,野球選手の野手を対象に体に装着したウェアラブルなセンサを用いて客観的な投球疲労の取得手法を提案する. Some baseball players have their shoulder and elbow injuries cause of the disorder of the throwing form and an overwork. Fielders especially practice without knowledge of their bodies and care for it. However it is difficult to recognize their throwing form and number of throws. This research has been supported player's performance by analysis of player's throwing using sanjigen-motion-capcher and douga-bunnseki for all time. But this research has problems with place and time. Because it doesn't assume actual practices. So we are researching for their new supports to improve their form and to take a break by using the radio small acceleration sensor which you can check your throwing form and number of throws easily. 収録刊行物

Image Generative Semantic Communication with Multi-Modal Similarity Estimation for Resource-Limited Networks

Eri Hosonuma, Taku Yamazaki, Takumi Miyoshi, Akihito Taya, Yuuki Nishiyama, Kaoru Sezaki

Estimating Sunlight Using GNSS Signal Strength from Smartphone

Yuuki Nishiyama, Kosuke Hatai, Kota Tsubouchi, Kaoru Sezaki

Enhancing Team-Level Behavior Change Using Information Sharing

Yuuki Nishiyama