Search

International audience; While past research has explored the effects of different targeted promotion strategies (e.g., temporal, geographical, crowdedness-based), it is yet to take into account the effect of social context (e.g., group cohesiveness) on consumer responses to such promotions. We develop a metric for measuring tie-strengths between individuals in a co-located group and compute a holistic measure, group cohesiveness, from longitudinal physical location data. We deploy this new metric in a field experiment to explore the effect of group cohesiveness (high vs. low) on consumer responses to mobile promotions of varying lead-times. Across all our studies, we demonstrate that shorter promotion lead-times are preferred more by individuals in the presence of high cohesive groups (versus low cohesive groups). Thus, marketers should take advantage of social context attributes (such as group cohesiveness), which can be derived from ubiquitous location tracking data to tailor their promotion based on consumers’ real-time activities.

Traditional mobility prediction literature focuses primarily on improved methods to extract latent patterns from individual-specific movement data. When such predictions are incorrect, we ascribe it to 'random' or 'unpredictable' changes in a user's movement behavior. Our hypothesis, however, is that such apparently-random deviations from daily movement patterns can, in fact, of ten be anticipated. In particular, we develop a methodology for predicting Likelihood of Future Non-Conformance (LFNC), based on two central hypotheses: (a) the likelihood of future deviations in movement behavior is positively correlated to the intensity of such trajectory deviations observed in the user's recent past, and (b) the likelihood of such future deviations increases if the user's strong-ties have also recently exhibited such non-conformant movement behavior. We use extensive longitudinal indoor location data (spanning 4+ months) from an urban university campus to validate these hypotheses, and then show that these features can be used to build an accurate non-conformance predictor: it can predict non-conformant mobility behavior two hours in advance with an AUC ≥ 0.85, significantly outperforming the baseline. We also show that this prediction methodology holds for a representative outdoor public-transport based mobility dataset. Finally, we use a real-world mobile crowd-sourcing application to show the practical impact of such non-conformance: failure to identify such likely anomalous movement behavior causes workers to suffer a noticeable drop in task completion rates and reduces the spatial spread of successfully completed tasks.

The paper discusses an emerging suite of machine intelligence services that are of increasing importance in the highly instrumented world of the Internet of Things (IoT). The suite, called Eugene, would offer a form of intelligent behavior (based on deep neural networks) to otherwise simple embedded devices; the clients of the service. These devices would benefit from service resources to learn from data and to perform intelligent inference, classification, prediction, and estimation tasks that they are too limited to carry out on their own. The paper discusses the taxonomy of such services and the state of implementation, as well as the various challenges entailed, including scheduling, caching (of intelligent functions), and cooperative learning.

The paper describes a vision for dependable application of machine learning-based inferencing on resource-constrained edge devices. The high computational overhead of sophisticated deep learning learning techniques imposes a prohibitive overhead, both in terms of energy consumption and sustainable processing throughput, on such resource-constrained edge devices (e.g., audio or video sensors). To overcome these limitations, we propose a “cognitive edge” paradigm, whereby (a) an edge device first autonomously uses statistical analysis to identify potential collaborative IoT nodes, and (b) the IoT nodes then perform real-time sharing of various intermediate state to improve their individual execution of machine intelligence tasks. We provide an example of such collaborative inferencing for an exemplar network of video sensors, showing how such collaboration can significantly improve accuracy, reduce latency and decrease communication bandwidth compared to non-collaborative baselines. We also identify various challenges in realizing such a cognitive edge, including the need to ensure that the inferencing tasks do not suffer catastrophically in the presence of malfunctioning peer devices. We then introduce the soon-to-be deployed Cognitive IoT testbed at SMU, explaining the various features that enable empirical testing of various novel edge-based ML algorithms.

While analysis of urban commuting data has a long and demonstrated history of providing useful insights into human mobility behavior, such analysis has been performed largely in offline fashion and to aid medium-to-long term urban planning. In this work, we demonstrate the power of applying predictive analytics on real-time mobility data, specifically the smart-card generated trip data of millions of public bus commuters in Singapore, to create two novel and "live" smart city services. The key analytical novelty in our work lies in combining two aspects of urban mobility: (a) conformity: which reflects the predictability in the aggregated flow of commuters along bus routes, and (b) regularity: which captures the repeated trip patterns of each individual commuter. We demonstrate that the fusion of these two measures of behavior can be performed at city-scale using our BuScope platform, and can be used to create two innovative smart city applications. The Last-Mile Demand Generator provides O(mins) lookahead into the number of disembarking passengers at neighborhood bus stops; it achieves over 85% accuracy in predicting such disembarkations by an ingenious combination of individual-level regularity with aggregate-level conformity. By moving driverless vehicles proactively to match this predicted demand, we can reduce wait times for disembarking passengers by over 75%. Independently, the Neighborhood Event Detector uses outlier measures of currently operating buses to detect and spatiotemporally localize dynamic urban events, as much as 1.5 hours in advance, with a localization error of 450 meters., Comment: ACM MobiSys 2019

Urban planners and economists alike have strong interest in understanding the inter-dependency of land use and people flow. The two-pronged problem entails systematic modeling and understanding of how land use impacts crowd flow to an area and in turn, how the influx of people to an area (or lack thereof) can influence the viability of business entities in that area. With cities becoming increasingly sensor-rich, for example, digitized payments for public transportation and constant trajectory tracking of buses and taxis, understanding and modelling crowd flows at the city scale, as well as, at finer granularity such as at the neighborhood level, has now become possible. Integrating such understanding with heterogeneous data such as land use profiles, demographics, and social media, enables richer studies on land use and its interdependence on mobility. In this work, we share findings from our preliminary efforts and identify key lines of research inquiry that can help urban planners towards data-driven policy decisions.

With the increased focus on making cities "smarter", we see an upsurge in investment in sensing technologies embedded in the urban infrastructure. The deployment of GPS sensors aboard taxis and buses, smartcards replacing paper tickets, and other similar initiatives have led to an abundance of data on human mobility, generated at scale and available real-time. Further still, users of social media platforms such as Twitter and LBSNs continue to voluntarily share multimedia content revealing in-situ information on their respective localities. The availability of such longitudinal multimodal data not only allows for both the characterization of the dynamics of the city, but also, in detecting anomalies, resulting from events (e.g., concerts) that disrupt such dynamics, transiently. In this work, we investigate the capabilities of such urban sensor modalities, both physical and social, in detecting a variety of local events of varying intensities (e.g., concerts) using statistical outlier detection techniques. We look at loading levels on arriving bus stops, telecommunication records, and taxi trips, accrued via the public APIs made available through the local transport authorities from Singapore and New York City, and Twitter/Foursquare check-ins collected during the same period, and evaluate against a set of events assimilated from multiple event websites. In particular, we report on our early findings on (1) the spatial impact evident via each modality (i.e., how far from the event venue is the anomaly still present), and (2) the utility in combining decisions from the collection of sensors using rudimentary fusion techniques.

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.

Singapore’s “smart city” agenda is driving the government to provide public access to a broader variety of urban informatics sources, such as images from traffic cameras and information about buses servicing different bus stops. Such informatics data serves as probes of evolving conditions at different spatiotemporal scales. This paper explores how such multi-modal informatics data can be used to establish the normal operating conditions at different city locations, and then apply appropriate outlier-based analysis techniques to identify anomalous events at these selected locations. We will introduce the overall architecture of sociophysical analytics, where such infrastructural data sources can be combined with social media analytics to not only detect such anomalous events, but also localize and explain them. Using the annual Formula-1 race as our candidate event, we demonstrate a key difference between the discriminative capabilities of different sensing modes: while social media streams provide discriminative signals during or prior to the occurrence of such an event, urban informatics data can often reveal patterns that have higher persistence, including before and after the event. In particular, we shall demonstrate how combining data from (i) publicly available Tweets, (ii) crowd levels aboard buses, and (iii) traffic cameras can help identify the Formula-1 driven anomalies, across different spatiotemporal boundaries.

The Android Open Source Project(AOSP) has seen tremendous traction over the past decade, and as such, the bug repository is growing in scale. With this growth, the effort required for project members to triage incoming new reports to identify whether it is a duplicate issue that has already been addressed, or receiving attention, is also on the rise. In this work, we create dataset of issues from the Android issue tracker, and use standard IR techniques such as VSM and LDA to understand their capability in such similar issue retrieval. Further, we combine VSM and LDA to evaluate its usefulness. We find that, overall, VSM performs better with this dataset.

Understanding one's group context in indoor spaces is useful for many reasons -- e.g., at a shopping mall, knowing a customer's group context can help in offering context-specific incentives, or estimating taxi demand for customers exiting the mall. Group detection and monitoring using WiFi-based indoor location traces fails when users are invisible (either because they don't carry smartphones, or because their WiFi is turned OFF) or when location tracking is inaccurate. In this paper, we propose a multi-modal group detection system that fuses two independent modes: video and WiFi, for detecting groups with low latency and high accuracy. We present preliminary results from a micro-study with 20 group episodes and report an overall precision of 0.81 and recall of 0.9, an improvement of over ≈20% over WiFi-based group detection.

In this paper, we present LiveLabs, a first-of-its-kind testbed that is deployed across a university campus, convention centre, and resort island and collects real-time attributes such as location, group context etc., from hundreds of opt-in participants. These venues, data, and participants are then made available for running rich human-centric behavioural experiments that could test new mobile sensing infrastructure, applications, analytics, or more social-science type hypotheses that influence and then observe actual user behaviour. We share case studies of how researchers from around the world have and are using LiveLabs, and our experiences and lessons learned from building, maintaining, and expanding Live-Labs over the last three years.

The paper explores the use of correlation across features extracted from different sensing channels to help in urban situational understanding. We use real-world datasets to show how such correlation can improve the accuracy of detection of city-wide events by combining metadata analysis with image analysis of Instagram content. We demonstrate this through a case study on the Singapore Haze. We show that simple ontological relationships and reasoning can significantly help in automating such correlation-based understanding of transient urban events.

In 2015, Singaporean have experienced one of the worse air pollution crises in history. With datasets from a well-known photo sharing social network, we analyze how this haze affects Singaporean's daily life. We will share our preliminary results in this paper.

The US Army Research Laboratory (ARL) and the Air Force Research Laboratory (AFRL) have established a collaborative research enterprise referred to as the Situational Understanding Research Institute (SURI). The goal is to develop an information processing framework to help the military obtain real-time situational awareness of physical events by harnessing the combined power of multiple sensing sources to obtain insights about events and their evolution. It is envisioned that one could use such information to predict behaviors of groups, be they local transient groups (e.g., Protests) or widespread, networked groups, and thus enable proactive prevention of nefarious activities. This paper presents a vision of how social media sources can be exploited in the above context to obtain insights about events, groups, and their evolution.

In this paper, we show that characterizing user behaviour from location or smartphone usage traces, without accounting for the interaction of individuals in physical-world groups, can lead to erroneous results. We conducted one of the largest studies in the UbiComp domain thus far, involving indoor location traces of more than 6,000 users, collected over a 4-month period at our university campus, and further studied fine-grained App usage of a subset of 156 Android users. We apply a state-of-the-art group detection algorithm to annotate such location traces with group vs. individual context, and then show that individuals vs. groups exhibit significant differences along three behavioural traits: (1) the mobility pattern, (2) the responsiveness to calls / SMSs and (3) application usage. We show that these significant differences are robust to underlying errors in the group detection technique and that the use of such group context leads to behavioural results that differ from those reported in prior popular work.

Growing interest in quantified self has led to the popularity of lifelogging applications. In particular, health and wellness related applications have seen an upsurge with the advent of wearables such as the Fitbit. In this paper, we focus on the quality of sleep that directly impacts the overall wellness of individuals. In particular, in this work, we present a first of its kind study that (1) unobtrusively quantifies the quality of sleep and (2) seeks to identify attributing aspects of our daily lives such as an individual's usage of apps throughout the day and his/her physical environment that may affect sleep. We use real life, in-situ smartphone data from 400+ undergraduate students over an observation period of 15 months, and present our initial observations.

Understanding the personality traits and current attitudes of individual consumers is crucial for retailers and mobile advertisers. In this paper, we investigate the phenomenon of "variety seeking tendencies" in mobile users in their (1) online (represented by their App usage behavior), and (2) physical (represented by their location visits) worlds. We show that different categories of users exhibit different levels of variety. Further, by analyzing at various time scales, we show that there exists correlation between when a person is likely to visit new places in the real world and when he/she is likely to explore new Apps in the online world.

In this paper, we argue for expanded research into an area called Socio-Physical Analytics, that focuses on combining the behavioral insight gained from mobile-sensing based monitoring of physical behavior with the inter-personal relationships and preferences deduced from online social networks. We highlight some of the research challenges in combining these heterogeneous data sources and then describe some examples of our ongoing work (based on real-world data being collected at SMU) that illustrate two aspects of socio-physical analytics: (a) how additional demographic and online analytics based attributes can potentially provide better insights into the preferences and behaviors of individuals or groups (in terms of movement prediction and understanding of physical vs. online interactions), and (b) how online and physical interactions can help us discover latent characteristics of physical spaces and entities.