Search

Cardiorespiratory fitness is an established predictor of metabolic disease and mortality. Fitness is directly measured as maximal oxygen consumption (VO$_{2}max$), or indirectly assessed using heart rate responses to standard exercise tests. However, such testing is costly and burdensome because it requires specialized equipment such as treadmills and oxygen masks, limiting its utility. Modern wearables capture dynamic real-world data which could improve fitness prediction. In this work, we design algorithms and models that convert raw wearable sensor data into cardiorespiratory fitness estimates. We validate these estimates' ability to capture fitness profiles in free-living conditions using the Fenland Study (N=11,059), along with its longitudinal cohort (N=2,675), and a third external cohort using the UK Biobank Validation Study (N=181) who underwent maximal VO$_{2}max$ testing, the gold standard measurement of fitness. Our results show that the combination of wearables and other biomarkers as inputs to neural networks yields a strong correlation to ground truth in a holdout sample (r = 0.82, 95CI 0.80-0.83), outperforming other approaches and models and detects fitness change over time (e.g., after 7 years). We also show how the model's latent space can be used for fitness-aware patient subtyping paving the way to scalable interventions and personalized trial recruitment. These results demonstrate the value of wearables for fitness estimation that today can be measured only with laboratory tests., Accepted in Nature Digital Medicine, 16 pages

Mental health plays a key role in everyone’s day-to-day lives, impacting our thoughts, behaviours, and emotions. Also, over the past years, given its ubiquitous and affordable characteristics, the use of smartphones and wearable devices has grown rapidly and provided support within all aspects of mental health research and care, spanning from screening and diagnosis to treatment and monitoring, and attained significant progress to improve remote mental health interventions. While there are still many challenges to be tackled in this emerging cross-discipline research field, such as data scarcity, lack of personalisation, and privacy concerns, it is of primary importance that innovative signal processing and deep learning techniques are exploited. Particularly, recent advances in deep learning can help provide the key enabling technology for the development of the next-generation user-centric mobile mental health applications. In this article, we first brief basic principles associated with mobile device-based mental health analysis, review the main system components, and highlight conventional technologies involved. Next, we describe several major challenges and various deep learning technologies that have potentials for a strong contribution in dealing with these challenges, respectively. Finally, we discuss other remaining problems which need to be addressed via research collaboration across multiple disciplines.

The PAM intervention is a behavioural intervention to support adherence to anti-hypertensive medications and therefore to lower blood pressure. This feasibility trial recruited 101 nonadherent patients (54% male, mean age 65.8 years) with hypertension and high blood pressure from nine general practices in the UK. The trial had 15.5% uptake and 7.9% attrition rate. Patients were randomly allocated to two groups: the intervention group (n = 61) received the PAM intervention as an adjunct to usual care; the control group (n = 40) received usual care only. At 3 months, biochemically validated medication adherence was improved by 20% (95% CI 3–36%) in the intervention than control, and systolic blood pressure was reduced by 9.16 mmHg (95% CI 5.69–12.64) in intervention than control. Improvements in medication adherence and reductions in blood pressure suggested potential intervention effectiveness. For a subsample of patients, improvements in medication adherence and reductions in full lipid profile (cholesterol 1.39 mmol/mol 95% CI 0.64–1.40) and in glycated haemoglobin (3.08 mmol/mol, 95% CI 0.42–5.73) favoured the intervention. A larger trial will obtain rigorous evidence about the potential clinical effectiveness and cost-effectiveness of the intervention.Trial registration Trial date of first registration 28/01/2019. ISRCTN74504989. https://doi.org/10.1186/ISRCTN74504989.

Recent work has shown the potential of using audio data (eg, cough, breathing, and voice) in the screening for COVID-19. However, these approaches only focus on one-off detection and detect the infection given the current audio sample, but do not monitor disease progression in COVID-19. Limited exploration has been put forward to continuously monitor COVID-19 progression, especially recovery, through longitudinal audio data. Tracking disease progression characteristics could lead to more timely treatment. The primary objective of this study is to explore the potential of longitudinal audio samples over time for COVID-19 progression prediction and, especially, recovery trend prediction using sequential deep learning techniques. Crowdsourced respiratory audio data, including breathing, cough, and voice samples, from 212 individuals over 5-385 days were analyzed. We developed a deep learning-enabled tracking tool using gated recurrent units (GRUs) to detect COVID-19 progression by exploring the audio dynamics of the individuals' historical audio biomarkers. The investigation comprised 2 parts: (1) COVID-19 detection in terms of positive and negative (healthy) tests, and (2) longitudinal disease progression prediction over time in terms of probability of positive tests. The strong performance for COVID-19 detection, yielding an AUROC of 0.79, a sensitivity of 0.75, and a specificity of 0.71 supported the effectiveness of the approach compared to methods that do not leverage longitudinal dynamics. We further examined the predicted disease progression trajectory, displaying high consistency with test results with a correlation of 0.75 in the test cohort and 0.86 in a subset of the test cohort who reported recovery. Our findings suggest that monitoring COVID-19 evolution via longitudinal audio data has potential in the tracking of individuals' disease progression and recovery., Comment: Updated title. Revised format according to journal requirements

Machine learning and deep learning have shown great promise in mobile sensing applications, including Human Activity Recognition. However, the performance of such models in real-world settings largely depends on the availability of large datasets that captures diverse behaviors. Recently, studies in computer vision and natural language processing have shown that leveraging massive amounts of unlabeled data enables performance on par with state-of-the-art supervised models. In this work, we present SelfHAR, a semi-supervised model that effectively learns to leverage unlabeled mobile sensing datasets to complement small labeled datasets. Our approach combines teacher-student self-training, which distills the knowledge of unlabeled and labeled datasets while allowing for data augmentation, and multi-task self-supervision, which learns robust signal-level representations by predicting distorted versions of the input. We evaluated SelfHAR on various HAR datasets and showed state-of-the-art performance over supervised and previous semi-supervised approaches, with up to 12% increase in F1 score using the same number of model parameters at inference. Furthermore, SelfHAR is data-efficient, reaching similar performance using up to 10 times less labeled data compared to supervised approaches. Our work not only achieves state-of-the-art performance in a diverse set of HAR datasets, but also sheds light on how pre-training tasks may affect downstream performance., Comment: Accepted for publication in Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies (IMWUT) 2021

Deep neural networks (DNNs) continue to demonstrate superior generalization performance in an increasing range of applications, including speech recognition and image understanding. Recent innovations in compression algorithms, design of efficient architectures and hardware accelerators have prompted a rapid growth in deploying DNNs on mobile and IoT devices to redefine user experiences. Relying on the superior inference quality of DNNs, various voice-enabled devices have started to pervade our everyday lives and are increasingly used for, e.g., opening and closing doors, starting or stopping washing machines, ordering products online, and authenticating monetary transactions. As the popularity of these voice-enabled services increases, so does their risk of being attacked. Recently, DNNs have been shown to be extremely brittle under adversarial attacks and people with malicious intentions can potentially exploit this vulnerability to compromise DNN-based voice-enabled systems. Although some existing work already highlights the vulnerability of audio models, very little is known of the behaviour of compressed on-device audio models under adversarial attacks. This paper bridges this gap by investigating thoroughly the vulnerabilities of compressed audio DNNs and makes a stride towards making compressed models robust. In particular, we propose a stochastic compression technique that generates compressed models with greater robustness to adversarial attacks. We present an extensive set of evaluations on adversarial vulnerability and robustness of DNNs in two diverse audio recognition tasks, while considering two popular attack algorithms: FGSM and PGD. We found that error rates of conventionally trained audio DNNs under attack can be as high as 100%. Under both white- and black-box attacks, our proposed approach is found to decrease the error rate of DNNs under attack by a large margin.

User authentication is key in user authorization on smart and personal devices. Over the years, several authentication mechanisms have been proposed: these also include behavioral-based biometrics. However, behavioral-based biometrics suffer from two issues: they are prone to degradation in performance (accuracy) over time (e.g., due to data distribution changes arising from user behavior) and the need to learn the machine learning model from scratch, when adding new users. In this paper, we propose ContAuth, a system that can enhance the robustness of behavioral-based authentication. ContAuth continuously adapts to new incoming data (data incremental learning) and is able to add new users without retraining (class incremental learning). Specifically, ContAuth combines deep learning models with online learning models to achieve learning on the fly, thereby preventing a severe drop in the accuracy between sessions (over time). To add new users, ContAuth employs class incremental learning methods. We evaluate ContAuth on multiple behavior-based user authentication modalities: breathing, gait. and EMG. Our results show that our framework can help True Positive Rate (TPR) to remain high (>85 %) compared to other methods for all the modalities except EMG (>70%) across the sessions while keeping False Positive Rates (FPR) at a minimum (0-10%). It can achieve up to 35% improvement in TPR over a traditional deep learning model. Additionally, iCaRL (an incremental learning method) enables ContAuth to allow the addition of new users by alleviating catastrophic forgetting, to a large extent. Finally, we also show that ContAuth can be deployed efficiently and effectively on device, further providing data privacy.

Continuous audio analysis from embedded and mobile devices is an increasingly important application domain. More and more, appliances like the Amazon Echo, along with smartphones and watches, and even research prototypes seek to perform multiple discriminative tasks simultaneously from ambient audio; for example, monitoring background sound classes (e.g., music or conversation), recognizing certain keywords (‘Hey Siri' or ‘Alexa'), or identifying the user and her emotion from speech. The use of deep learning algorithms typically provides state-of-the-art model performances for such general audio tasks. However, the large computational demands of deep learning models are at odds with the limited processing, energy and memory resources of mobile, embedded and IoT devices. In this paper, we propose and evaluate a novel deep learning modeling and optimization framework that specifically targets this category of embedded audio sensing tasks. Although the supported tasks are simpler than the task of speech recognition, this framework aims at maintaining accuracies in predictions while minimizing the overall processor resource footprint. The proposed model is grounded in multi-task learning principles to train shared deep layers and exploits, as input layer, only statistical summaries of audio filter banks to further lower computations. We find that for embedded audio sensing tasks our framework is able to maintain similar accuracies, which are observed in comparable deep architectures that use single-task learning and typically more complex input layers. Most importantly, on an average, this approach provides almost a 2.1× reduction in runtime, energy, and memory for four separate audio sensing tasks, assuming a variety of task combinations.

A recent trend in corporate real-estate is Activity-Based Working (ABW). The ABW concept removes designated desks but offers different work settings designed to support typical work activities. In this context there is still a need for objective data to understand the implications of these design decisions. We aim to contribute by using automated data collection to study how ABW’s principles impact office usage and dynamics. To this aim we analyse team dynamics and employees’ tie strength in relation to space usage and organisational hierarchy using data collected with wearable devices in a company adopting ABW principles. Our findings show that the office fosters interactions across team boundaries and among the lower levels of the hierarchy suggesting a strong lateral communication. Employees also tend to have low space exploration on a daily basis which is instead more prevalent during an average week and strong social clusters seem to be resisting the ABW principles of space dynamics. With the availability of two additional data sets about social encounters in traditional offices we highlight traits emerging from the application of ABW’s principles. In particular, we observe how the absence of designated desks might be responsible for more rapid dynamics inside the office. In more general terms, this work opens the door to new and scalable technology-based methodologies to study dynamic office usage and social interactions.

Understanding and predicting human mobility is vital to a large number of applications, ranging from recommendations to safety and urban service planning. In some travel applications, the ability to accurately predict the user's future trajectory is vital for delivering high quality of service. The accurate prediction of detailed trajectories would empower location-based service providers with the ability to deliver more precise recommendations to users. Existing work on human mobility prediction has mainly focused on the prediction of the next location (or the set of locations) visited by the user, rather than on the prediction of the continuous trajectory (sequences of further locations and the corresponding arrival and departure times). Furthermore, existing approaches often return predicted locations as regions with coarse granularity rather than geographical coordinates, which limits the practicality of the prediction. In this paper, we introduce a novel trajectory prediction problem: given historical data and a user's initial trajectory in the morning, can we predict the user's full trajectory later in the day (e.g. the afternoon trajectory)? The predicted continuous trajectory includes the sequence of future locations, the stay times, and the departure times. We first conduct a comprehensive analysis about the relationship between morning trajectories and the corresponding afternoon trajectories, and found there is a positive correlation between them. Our proposed method combines similarity metrics over the extracted temporal sequences of locations to estimate similar informative segments across user trajectories. Our evaluation shows results on both labeled and geographical trajectories with a prediction error reduced by 10-35% in comparison to the baselines. This improvement has the potential to enable precise location services, raising usefulness to users to unprecedented levels. We also present empirical evaluations with Markov model and Long Short Term Memory (LSTM), a state-of-the-art Recurrent Neural Network model. Our proposed method is shown to be more effective when smaller number of samples are used and is exponentially more efficient than LSTM.

Estimating revenue and business demand of a newly opened venue is paramount as these early stages often involve critical decisions such as first rounds of staffing and resource allocation. Traditionally, this estimation has been performed through coarse-grained measures such as observing numbers in local venues or venues at similar places (e.g., coffee shops around another station in the same city). The advent of crowdsourced data from devices and services carried by individuals on a daily basis has opened up the possibility of performing better predictions of temporal visitation patterns for locations and venues. In this paper, using mobility data from Foursquare, a location-centric platform, we treat venue categories as proxies for urban activities and analyze how they become popular over time. The main contribution of this work is a prediction framework able to use characteristic temporal signatures of places together with k-nearest neighbor metrics capturing similarities among urban regions, to forecast weekly popularity dynamics of a new venue establishment in a city neighborhood. We further show how we are able to forecast the popularity of the new venue after one month following its opening by using locality and temporal similarity as features. For the evaluation of our approach we focus on London. We show that temporally similar areas of the city can be successfully used as inputs of predictions of the visit patterns of new venues, with an improvement of 41% compared to a random selection of wards as a training set for the prediction task. We apply these concepts of temporally similar areas and locality to the real-time predictions related to new venues and show that these features can effectively be used to predict the future trends of a venue. Our findings have the potential to impact the design of location-based technologies and decisions made by new business owners., Foursquare

Human mobility is often represented as a mobility network, or graph, with nodes representing places of significance which an individual visits, such as their home, work, places of social amenity, etc., and edge weights corresponding to probability estimates of movements between these places. Previous research has shown that individuals can be identified by a small number of geolocated nodes in their mobility network, rendering mobility trace anonymization a hard task. In this paper we build on prior work and demonstrate that even when all location and timestamp information is removed from nodes, the graph topology of an individual mobility network itself is often uniquely identifying. Further, we observe that a mobility network is often unique, even when only a small number of the most popular nodes and edges are considered. We evaluate our approach using a large dataset of cell-tower location traces from 1 500 smartphone handsets with a mean duration of 430 days. We process the data to derive the top−N places visited by the device in the trace, and find that 93% of traces have a unique top−10 mobility network, and all traces are unique when considering top−15 mobility networks. Since mobility patterns, and therefore mobility networks for an individual, vary over time, we use graph kernel distance functions, to determine whether two mobility networks, taken at different points in time, represent the same individual. We then show that our distance metrics, while imperfect predictors, perform significantly better than a random strategy and therefore our approach represents a significant loss in privacy.

Mobile devices are becoming pervasive to our daily lives: they follow us everywhere and we use them for much more than just communication. These devices are also equipped with a myriad of different sensors that have the potential to allow the tracking of human activities, user patterns, location, direction and much more. Following this direction, many movements including sports, quantified self, and mobile health ones are starting to heavily rely on this technology, making it pivotal that the sensors offer high accuracy. However, heterogeneity in hardware manufacturing, slight substrate differences, electronic interference as well as external disturbances are just few of the reasons that limit sensor output accuracy which in turn hinders sensor usage in applications which need very high granularity and precision, such as quantified-self applications. Although, calibration of sensors is a widely studied topic in literature to the best of our knowledge no publicly available research exists that specifically tackles the calibration of mobile phones and existing methods that can be adapted for use in mobile devices not only require user interaction but they are also not adaptive to changes. Additionally, alternative approaches for performing more granular and accurate sensing exploit body-wide sensor networks using mobile phones and additional sensors; as one can imagine these techniques can be bulky, tedious, and not particularly user friendly. Moreover, existing techniques for performing data corrections post-acquisition can produce inconsistent results as they miss important context information provided from the device itself; which when used, has been shown to produce better results without a imposing a significant power-penalty. In this paper we introduce a novel multiposition calibration scheme that is specifically targeted at mobile devices Our scheme exploits machine learning techniques to perform an adaptive, power-efficient auto-calibration procedure with which achieves high output sensor accuracy when compared to state of the art techniques without requiring any user interaction or special equipment beyond device itself Moreover, the energy costs associated with our approach are lower than the alternatives (such as Kalman filter based solutions) and the overall power penalty is < 5% when compared against power usage that is exhibited when using uncalibrated traces, thus, enabling our technique to be used efficiently on a wide variety of devices Finally, our evaluation illustrates that calibrated signals offer a tangible benefit in classification accuracy, ranging from 3 to 10%, over uncalibrated ones when using state of the art classifiers, on the other hand when using simpler SVM classifiers the classification improvement is boosted ranging from 8% to 12% making lower performing classifiers much more reliable Additionally, we show that for similar activities which are hard to distinguish otherwise, we reach an accuracy of > 95% when using neural network classifiers and > 88% when using SVM classifiers where uncalibrated data classification only reaches ~ 85% and ~ 80% respectively This can be a make or break factor in the use of accelerometer and gyroscope data in applications requiring high accuracy e g sports, health, games and others

Wearables and mobile devices see the world through the lens of half a dozen low-power sensors, such as, barometers, accelerometers, microphones and proximity detectors. But differences between sensors ranging from sampling rates, discrete and continuous data or even the data type itself make principled approaches to integrating these streams challenging. How, for example, is barometric pressure best combined with an audio sample to infer if a user is in a car, plane or bike? Critically for applications, how successfully sensor devices are able to maximize the information contained across these multi-modal sensor streams often dictates the fidelity at which they can track user behaviors and context changes. This paper studies the benefits of adopting deep learning algorithms for interpreting user activity and context as captured by multi-sensor systems. Specifically, we focus on four variations of deep neural networks that are based either on fully-connected Deep Neural Networks (DNNs) or Convolutional Neural Networks (CNNs). Two of these architectures follow conventional deep models by performing feature representation learning from a concatenation of sensor types. This classic approach is contrasted with a promising deep model variant characterized by modality-specific partitions of the architecture to maximize intra-modality learning. Our exploration represents the first time these architectures have been evaluated for multimodal deep learning under wearable data -- and for convolutional layers within this architecture, it represents a novel architecture entirely. Experiments show these generic multimodal neural network models compete well with a rich variety of conventional hand-designed shallow methods (including feature extraction and classifier construction) and task-specific modeling pipelines, across a wide-range of sensor types and inference tasks (four different datasets). Although the training and inference overhead of these multimodal deep approaches is in some cases appreciable, we also demonstrate the feasibility of on-device mobile and wearable execution is not a barrier to adoption. This study is carefully constructed to focus on multimodal aspects of wearable data modeling for deep learning by providing a wide range of empirical observations, which we expect to have considerable value in the community. We summarize our observations into a series of practitioner rules-of-thumb and lessons learned that can guide the usage of multimodal deep learning for activity and context detection.

Online social networking services entice millions of users to spend hours every day interacting with each other. The focus of this work is to explain the effect that geographic distance has on online social interactions and, simultaneously, to understand the interplay between the social characteristics of friendship ties and their spatial properties. We analyze data from a large-scale online social network, Tuenti, with about 10 million active users: our sample includes user profiles, user home locations and online social interactions among Tuenti members. Our findings support the idea that spatial distance constraints whom users interact with, but not the intensity of their social interactions. Furthermore, friendship ties belonging to denser connected groups tend to arise at shorter spatial distances than social ties established between members belonging to different groups. Finally, we show that our findings mostly do not depend on the age of the users, although younger users seem to be slightly more constrained to shorter geographic distances. Augmenting social structure with geographic information adds a new dimension to social network analysis and a large number of theoretical investigations and practical applications can be pursued for online social systems, with many promising outcomes. As the amount of available location-based data is increasing, our findings and results open the door to future possibilities: researchers would benefit from these insights when studying online social services, while developers should be aware of these additional possibilities when building systems and applications related to online social platforms.

Economic and urban planning agencies have strong interest in tackling the hard problem of predicting the odds of survival of individual retail businesses. In this work, we tap urban mobility data available both from a location-based intelligence platform, Foursquare, and from public transportation agencies, and investigate whether mobility-derived features can help foretell the failure of such retail businesses, over a 6 month horizon, across 10 distinct cities spanning the globe. We hypothesise that the survival of such a retail outlet is correlated with not only venue-specific characteristics but also broader neighbourhood-level effects. Through careful statistical analysis of Foursquare and taxi mobility data, we uncover a set of discriminative features, belonging to the neighbourhood's static characteristics, the venue-specific customer visit dynamics, and the neighbourhood's mobility dynamics. We demonstrate that classifiers trained on such features can predict such survival with high accuracy, achieving approximately 80% precision and recall across the cities. We also show that the impact of such features varies across new and established venues and across different cities. Besides achieving a significant improvement over past work on business vitality prediction, our work demonstrates the vital role that mobility dynamics plays in the economic evolution of a city.

Bicycle sharing systems are increasingly being deployed in urban areas around the world, alongside online maps that disclose the state (i.e., location, number of bicycles/number of free parking slots) of stations in each city. Recent work has demonstrated how regularly monitoring these online maps allows for a granular analysis of a city’s cycling trends; further, the literature indicates that different cities have unique spatio-temporal patterns, reducing the generalisability of any insights or models derived from a single system. In this work, we analyse 4.5 months of online bike-sharing map data from ten cities which, combined, have 996 stations. While an aggregate comparison supports the view of cities having unique usage patterns, results of applying unsupervised learning to the temporal data shows that, instead, only the larger systems display heterogeneous behaviour, indicating that many of these systems share intrinsic similarities. We further show how these similarities are reflected in the predictability of stations’ occupancy data via a cross-city comparison of the error that a variety of approaches achieve when forecasting the number of bicycles that a station will have in the near future. We close by discussing the impact of uncovering these similarities on how future bicycle sharing systems can be designed, built, and managed.

User interaction patterns with mobile apps and notifications are generally complex due to the many factors involved. However a deep understanding of what influences them can lead to more acceptable applications that are able to deliver information at the right time. In this paper, we present for the first time an in-depth analysis of interaction behavior with notifications in relation to the location and activity of users. We conducted an in-situ study for a period of two weeks to collect more than 36,000 notifications, 17,000 instances of application usage, 77,000 location samples, and 487 days of daily activity entries from 26 students at a UK university. Our results show that users’ attention towards new notifications and willingness to accept them are strongly linked to the location they are in and in minor part to their current activity. We consider both users’ receptivity and attentiveness, and we show that different response behaviors are associated to different locations. These findings are fundamental from a design perspective since they allow us to understand how certain types of places are linked to specific types of interaction behavior. This information can be used as a basis for the development of novel intelligent mobile applications and services.

Measuring mental well-being with mobile sensing has been an increasingly active research topic. Pervasiveness of smartphones combined with the convenience of mobile app distribution platforms (e.g., Google Play) provide a tremendous opportunity to reach out to millions of users. However, the studies at the confluence of mental health and mobile sensing have been longitudinally limited, controlled, or confined to a small number of participants. In this paper we report on what we believe is the largest longitudinal in-the-wild study of mood through smartphones. We describe an Android app to collect participants’ self-reported moods and system triggered experience sampling data while passively measuring their physical activity, sociability, and mobility via their device’s sensors. We report the results of a large-scale analysis of the data collected for about three years from $\sim$ 18; 000 users. The paper makes three primary contributions. First, we show how we used physical and software sensors in smartphones to automatically and accurately identify routines. Then, we demonstrate the strong correlation between these routines and users’ personality, well-being perception, and other psychological variables. Finally, we explore predictability of users’ mood using their passive sensing data. Our findings show that, especially for weekends, mobile sensing can be used to predict users’ mood with an accuracy of about 70%. These results have the potential to impact the design of future mobile apps for mood/behavior tracking and interventions.

Physical activity, both exercise and non-exercise, has far-reaching benefits to physical health. Although exercise has also been linked to psychological health (e.g., happiness), little research has examined physical activity more broadly, taking into account non-exercise activity as well as exercise. We examined the relationship between physical activity (measured broadly) and happiness using a smartphone application. This app has collected self-reports of happiness and physical activity from over ten thousand participants, while passively gathering information about physical activity from the accelerometers on users' phones. The findings reveal that individuals who are more physically active are happier. Further, individuals are happier in the moments when they are more physically active. These results emerged when assessing activity subjectively, via self-report, or objectively, via participants' smartphone accelerometers. Overall, this research suggests that not only exercise but also non-exercise physical activity is related to happiness. This research further demonstrates how smartphones can be used to collect large-scale data to examine psychological, behavioral, and health-related phenomena as they naturally occur in everyday life.

The ability to detect and distinguish interactions in the workplace can shed light over productivity, team work and on employees’ use of space. Questionnaires and direct observations have often been used as mechanisms to identify office based interactions, however, these are either very time consuming, yield coarse grained information or do not scale to large numbers of people. Technology has been recently employed to cut costs and improve output, however precise interaction dynamics gathering often requires individuals to wear custom hardware. In this paper, we present an extensive evaluation of Bluetooth Low Energy (BLE) as a technology to monitor people proximity in the workplace. We examine the key parameters that affect the accuracy of the detected contacts and their impact on power consumption. We study how this system can be implemented on popular wearable devices (i.e., Android Wear and Tizen) and the resulting limitations. Through a real world deployment in a commercial organisation with 25 participants we evaluate the performances of a BLE-based proximity detection technique. Our results show the suitability of BLE for workplace inter- action detection and give guidance to vendors and Operating System (OS) developers on the impact of the restrictions regard- ing the use of BLE on commodity wearables.

BackgroundA major cause of lapse and relapse to smoking during a quit attempt is craving triggered by cues from a smoker's immediate environment. To help smokers address these cue-induced cravings when attempting to quit, we have developed a context-aware smoking cessation app, Q Sense, which uses a smoking episode-reporting system combined with location sensing and geofencing to tailor support content and trigger support delivery in real time. ObjectiveWe sought to (1) assess smokers’ compliance with reporting their smoking in real time and identify reasons for noncompliance, (2) assess the app's accuracy in identifying user-specific high-risk locations for smoking, (3) explore the feasibility and user perspective of geofence-triggered support, and (4) identify any technological issues or privacy concerns. MethodsAn explanatory sequential mixed-methods design was used, where data collected by the app informed semistructured interviews. Participants were smokers who owned an Android mobile phone and were willing to set a quit date within one month (N=15). App data included smoking reports with context information and geolocation, end-of-day (EoD) surveys of smoking beliefs and behavior, support message ratings, and app interaction data. Interviews were undertaken and analyzed thematically (N=13). Quantitative and qualitative data were analyzed separately and findings presented sequentially. ResultsOut of 15 participants, 3 (20%) discontinued use of the app prematurely. Pre-quit date, the mean number of smoking reports received was 37.8 (SD 21.2) per participant, or 2.0 (SD 2.2) per day per participant. EoD surveys indicated that participants underreported smoking on at least 56.2% of days. Geolocation was collected in 97.0% of smoking reports with a mean accuracy of 31.6 (SD 16.8) meters. A total of 5 out of 9 (56%) eligible participants received geofence-triggered support. Interaction data indicated that 50.0% (137/274) of geofence-triggered message notifications were tapped within 30 minutes of being generated, resulting in delivery of a support message, and 78.2% (158/202) of delivered messages were rated by participants. Qualitative findings identified multiple reasons for noncompliance in reporting smoking, most notably due to environmental constraints and forgetting. Participants verified the app’s identification of their smoking locations, were largely positive about the value of geofence-triggered support, and had no privacy concerns about the data collected by the app. ConclusionsUser-initiated self-report is feasible for training a cessation app about an individual’s smoking behavior, although underreporting is likely. Geofencing was a reliable and accurate method of identifying smoking locations, and geofence-triggered support was regarded positively by participants.

Does personality predict how people feel in different types of situations? The present research addressed this question using data from several thousand individuals who used a mood tracking smartphone application for several weeks. Results from our analyses indicated that people’s momentary affect was linked to their location, and provided preliminary evidence that the relationship between state affect and location might be moderated by personality. The results highlight the importance of looking at person-situation relationships at both the trait- and state-levels and also demonstrate how smartphones can be used to collect person and situation information as people go about their everyday lives.

Every day, people are reliant on smartphones to stay connected, organized, and entertained. Every year, smartphones are equipped with more processing power, digital storage, and sensors. As a result of their growing ubiquity and technical advances, there are now exciting opportunities for leveraging these devices in clinical health research, care, and intervention. Collectively, these opportunities are often referred to as the emergent “mobile health” or mHealth trend. Here, we briefly discuss 3 areas where smartphones can make a significant impact (ambulatory assessment, behavior monitoring, understanding and predicting outcomes), using a system that we have developed—Emotion Sense—as a case study. We also caution about some of the limitations of this technology.

Uber has recently been introducing novel practices in urban taxi transport. Journey prices can change dynamically in almost real time and also vary geographically from one area to another in a city, a strategy known as surge pricing. In this paper, we explore the power of the new generation of open datasets towards understanding the impact of the new disruption technologies that emerge in the area of public transport. With our primary goal being a more transparent economic landscape for urban commuters, we provide a direct price comparison between Uber and the Yellow Cab company in New York. We discover that Uber, despite its lower standard pricing rates, effectively charges higher fares on average, especially during short in length, but frequent in occurrence, taxi journeys. Building on this insight, we develop a smartphone application, OpenStreetCab, that offers a personalized consultation to mobile users on which taxi provider is cheaper for their journey. Almost five months after its launch, the app has attracted more than three thousand users in a single city. Their journey queries have provided additional insights on the potential savings similar technologies can have for urban commuters, with a highlight being that on average, a user in New York saves 6 U.S. Dollars per taxi journey if they pick the cheapest taxi provider. We run extensive experiments to show how Uber’s surge pricing is the driving factor of higher journey prices and therefore higher potential savings for our application’s users. Finally, motivated by the observation that Uber’s surge pricing is occurring more frequently that intuitively expected, we formulate a prediction task where the aim becomes to predict a geographic area’s tendency to surge. Using exogenous to Uber data, in particular Yellow Cab and Foursquare data, we show how it is possible to estimate customer demand within an area, and by extension surge pricing, with high accuracy.

While there is significant work on sensing and recognition of significant places for users, little attention has been given to users' significant routes. Recognizing these routine journeys, opens doors to the development of novel applications, like personalized travel alerts, and enhancement of user's travel experience. However, the high energy consumption of traditional location sensing technologies, such as GPS or WiFi based localization, is a barrier to passive and ubiquitous route sensing through smartphones. In this paper, we present a passive route sensing framework that continuously monitors a vehicle user solely through a phone's gyroscope and accelerometer. This approach can differentiate and recognize various routes taken by the user by time warping angular speeds experienced by the phone while in transit and is independent of phone orientation and location within the vehicle, small detours and traffic conditions. We compare the route learning and recognition capabilities of this approach with GPS trajectory analysis and show that it achieves similar performance. Moreover, with an embedded co-processor, common to most new generation phones, it achieves energy savings of an order of magnitude over the GPS sensor., ACM Conference on Embedded Networked Sensor Systems (SenSys 2014)

The layouts of the buildings we live in shape our everyday lives. In office environments, building spaces affect employees' communication, which is crucial for productivity and innovation. However, accurate measurement of how spatial layouts affect interactions is a major challenge and traditional techniques may not give an objective view.We measure the impact of building spaces on social interactions using wearable sensing devices. We study a single organization that moved between two different buildings, affording a unique opportunity to examine how space alone can affect interactions. The analysis is based on two large scale deployments of wireless sensing technologies: short-range, lightweight RFID tags capable of detecting face-to-face interactions. We analyze the traces to study the impact of the building change on social behavior, which represents a first example of using ubiquitous sensing technology to study how the physical design of two workplaces combines with organizational structure to shape contact patterns., 14 pages, 9 figures. To appear in ACM International Joint Conference on Pervasive and Ubiquitous Computing (Ubicomp 2014)

Understanding the social and behavioral forces behind event participation is not only interesting from the viewpoint of social science, but also has important applications in the design of personalized event recommender systems.This paper takes advantage of data from a widely used location-based social network, Foursquare, to analyze event patterns in three metropolitan cities. We put forward several hypotheses on the motivating factors of user participation and confirm that social aspects play a major role in determining the likelihood of a user to participate in an event. While an explicit social filtering signal accounting for whether friends are attending dominates the factors, the popularity of an event proves to also be a strong attractor. Further, we capture an implicit social signal by performing random walks in a high dimensional graph that encodes the place type preferences of friends and that proves especially suited to identify relevant niche events for users. Our findings on the extent to which the various temporal, spatial and social aspects underlie users' event preferences lead us to further hypothesize that a combination of factors better models users' event interests. We verify this through a supervised learning framework. We show that for one in three users in London and one in five users in New York and Chicago it identifies the exact event the user would attend among the pool of suggestions.

The rapidly growing adoption of sensor-enabled smartphones has greatly fueled the proliferation of applications that use phone sensors to monitor user behavior. A central sensor among these is the microphone which enables, for instance, the detection of valence in speech, or the identification of speakers. Deploying multiple of these applications on a mobile device to continuously monitor the audio environment allows for the acquisition of a diverse range of sound-related contextual inferences. However, the cumulative processing burden critically impacts the phone battery. To address this problem, we propose DSP.Ear - an integrated sensing system that takes advantage of the latest low-power DSP co-processor technology in commodity mobile devices to enable the continuous and simultaneous operation of multiple established algorithms that perform complex audio inferences. The system extracts emotions from voice, estimates the number of people in a room, identifies the speakers, and detects commonly found ambient sounds, while critically incurring little overhead to the device battery. This is achieved through a series of pipeline optimizations that allow the computation to remain largely on the DSP. Through detailed evaluation of our prototype implementation we show that, by exploiting a smartphone's co-processor, DSP.Ear achieves a 3 to 7 times increase in the battery lifetime compared to a solution that uses only the phone's main processor. In addition, DSP.Ear is 2 to 3 times more power efficient than a naive DSP solution without optimizations. We further analyze a large-scale dataset from 1320 Android users to show that in about 80-90% of the daily usage instances DSP.Ear is able to sustain a full day of operation (even in the presence of other smartphone workloads) with a single battery charge., Comment: 15 pages, 12th ACM Conference on Embedded Network Sensor Systems (SenSys '14)

We analyze two large datasets from technological networks with location and social data: user location records from an online location-based social networking service, and anonymized telecommunications data from a European cellphone operator, in order to investigate the differences between individual and group behavior with respect to physical location. We discover agreements between the two datasets: firstly, that individuals are more likely to meet with one friend at a place they have not visited before, but tend to meet at familiar locations when with a larger group. We also find that groups of individuals are more likely to meet at places that their other friends have visited, and that the type of a place strongly affects the propensity for groups to meet there. These differences between group and solo mobility has potential technological applications, for example, in venue recommendation in location-based social networks., 7 pages, 8 figures. Accepted for publication in PLOS One

The Olympic Games are an important sporting event with notable consequences for the general economic landscape of the host city. Traditional economic assessments focus on the aggregated impact of the event on the national income, but fail to provide micro-scale insights on why local businesses will benefit from the increased activity during the Games.In this paper we provide a novel approach to modeling the impact of the Olympic Games on local retailers by analyzing a dataset mined from a large location-based social service, Foursquare. We hypothesize that the spatial positioning of businesses as well as the mobility trends of visitors are primary indicators of whether retailers will rise their popularity during the event. To confirm this we formulate a retail winners prediction task in the context of which we evaluate a set of geographic and mobility metrics. We find that the proximity to stadiums, the diversity of activity in the neighborhood, the nearby area sociability, as well as the probability of customer flows from and to event places such as stadiums and parks are all vital factors. Through supervised learning techniques we demonstrate that the success of businesses hinges on a combination of both geographic and mobility factors. Our results suggest that location-based social networks, where crowdsourced information about the dynamic interaction of users with urban spaces becomes publicly available, present an alternative medium to assess the economic impact of large scale events in a city.

In many work environments, serendipitous interactions between members of different groups may lead to enhanced productivity, collaboration and knowledge dissemination. Two factors that may have an influence on such interactions are cultural differences between individuals in highly multicultural workplaces, and the layout and physical spaces of the workplace itself. In this work, we investigate how these two factors may facilitate or hinder inter-group interactions in the workplace. We analyze traces collected using wearable electronic badges to capture face-to-face interactions and mobility patterns of employees in a research laboratory in the UK. We observe that those who interact with people of different roles tend to come from collectivist cultures that value relationships and where people tend to be comfortable with social hierarchies, and that some locations in particular are more likely to host serendipitous interactions, knowledge that could be used by organizations to enhance communication and productivity., Comment: 10 pages, 4 figures. ACM CSCW 2014

In recent decades, the world has experienced rates of urban growth unparalleled in any other period of history and this growth is shaping the environment in which an increasing proportion of us live. In this paper, we use a longitudinal dataset from Foursquare, a location-based social network, to analyse urban growth across 100 major cities worldwide. Initially, we explore how urban growth differs in cities across the world. We show that there exists a strong spatial correlation, with nearby pairs of cities more likely to share similar growth profiles than remote pairs of cities. Subsequently, we investigate how growth varies inside cities and demonstrate that, given the existing local density of places, higher-than-expected growth is highly localized while lower-than-expected growth is more diffuse. Finally, we attempt to use the dataset to characterize competition between new and existing venues. By defining a measure based on the change in throughput of a venue before and after the opening of a new nearby venue, we demonstrate which venue types have a positive effect on venues of the same type and which have a negative effect. For example, our analysis confirms the hypothesis that there is large degree of competition between bookstores, in the sense that existing bookstores normally experience a notable drop in footfall after a new bookstore opens nearby. Other place types, such as museums, are shown to have a cooperative effect and their presence fosters higher traffic volumes to nearby places of the same type.