Longbiao Chen

NationTelescope: Monitoring and Visualizing Large-Scale Collective Behavior in LBSNs

Abstract The research of collective behavior has attracted a lot of attention in recent years, which can empower various applications, such as recommendation systems and intelligent transportation systems. However, in traditional social science, it is practically difficult to collect large-scale user behavior data. Fortunately, with the ubiquity of smartphones and Location Based Social Networks (LBSNs), users continuously report their activities online, which massively reflect their collective behavior. In this paper, we propose NationTelescope, a platform that monitors, compares and visualizes large-scale nation-wide user behavior in LBSNs. First, it continuously collects user behavior data from LBSNs. Second, it automatically generates behavior data summary and integrates an interactive map interface for data visualization. Third, in order to compare and visualize the behavioral differences across countries, it detects the discriminative activities according to the related traffic patterns in different countries. By implementing a prototype of NationTelescope platform, we evaluate its effectiveness and usability via two case studies and a system usability scale survey. The results show that the platform can not only efficiently capture, compare and visualize nation-wide collective behavior, but also achieve good usability and user experience.

Dynamic cluster-based over-demand prediction in bike sharing systems

Bike sharing is booming globally as a green transportation mode, but the occurrence of over-demand stations that have no bikes or docks available greatly affects user experiences. Directly predicting individual over-demand stations to carry out preventive measures is difficult, since the bike usage pattern of a station is highly dynamic and context dependent. In addition, the fact that bike usage pattern is affected not only by common contextual factors (e.g., time and weather) but also by opportunistic contextual factors (e.g., social and traffic events) poses a great challenge. To address these issues, we propose a dynamic cluster-based framework for over-demand prediction. Depending on the context, we construct a weighted correlation network to model the relationship among bike stations, and dynamically group neighboring stations with similar bike usage patterns into clusters. We then adopt Monte Carlo simulation to predict the over-demand probability of each cluster. Evaluation results using real-world data from New York City and Washington, D.C. show that our framework accurately predicts over-demand clusters and outperforms the baseline methods significantly.

Touch-driven interaction via an NFC-enabled smartphone

With the advent of smart home technologies, convenient and intuitive cyber-physical interaction methods are in great need. We demonstrate a touch-driven interaction system with home appliances via an NFC-enabled smartphone. By touching the phone to NFC tagged devices, we can establish connection and exchange information, control devices from phone, and share media between them. We also demonstrate three applications, Touch&Connect, Touch&Listen, and Touch&Watch, to demonstrate touch-driven interaction in a smart home environment.

Touch-Driven Interaction between Physical Space and Cyberspace with NFC

Nowadays, natural and intuitive interactions between physical space and cyberspace have become increasingly important in our daily life. In this paper, we propose touch-driven interaction, which translates touch action into information flow in cyberspace, and result in media activity in physical space. Using NFC technology, user can simply touch smart phone with NFC tags to interact with TV, stereo, digital frame, etc. Compared with recognition-based interactions, touch-driven interaction features full user controllability, fine-grain accuracy and high usability. We also present several applications, including Touch&Connect, Touch&Watch, Touch&Listen and Touch&MakeFriends, to demonstrate touch-driven interaction, and conclude that such interaction is both intuitive and convenient in daily life.

Sensing the Pulse of Urban Activity Centers Leveraging Bike Sharing Open Data

Understanding social activities in Urban Activity Centers can benefit both urban authorities and citizens. Traditionally, monitoring large social activities usually incurs significant costs of human labor and time. Fortunately, with the recent booming of urban open data, a wide variety of human digital footprints have become openly accessible, providing us with new opportunities to understand the social dynamics in the cities. In this paper, we resort to urban open data from bike sharing systems, and propose a two-phase framework to identify social activities in Urban Activity Centers based on bike sharing open data. More specifically, we first detect bike usage anomalies from the bike trip data, and then identify the potential social activities from the detected anomalies using a proposed heuristic method by considering both spatial and temporal constraints. We evaluate our framework based on the large-scale real-world dataset collected from the bike sharing system of Washington, D.C. The results show that our framework can efficiently identify social activities in different types of Urban Activity Centers and outperforms the baseline approach. In particular, our framework can identify 89% of the social activities in the major Urban Activity Centers of Washington, D.C.

Fine-Grained Urban Event Detection and Characterization Based on Tensor Cofactorization

Understanding the irregular crowd movement and social activities caused by urban events such as city festivals and concerts can benefit event management and city planning. Although various urban data can be exploited to detect such irregularities, the crowd mobility data (e.g., bike trip records) are usually in a mixed state with several basic patterns (e.g., eating, working, and recreation), making it difficult to separate concurrent events happening in the same region. The social activity data (e.g., social network check-ins) are usually oversparse, hindering the fine-grained characterization of urban events. In this paper, we propose a tensor cofactorization-based data fusion framework for fine-grained urban event detection and characterization leveraging crowd mobility data and social activity data. First, we adopt a nonnegative tensor cofactorization approach to decompose the crowd mobility tensor into several basic patterns, with the help of the auxiliary social activity tensor. We then use a multivariate-outlier-detection-based method to identify irregularities from the decomposed basic patterns and aggregate them to detect and characterize the associated urban events. We evaluate the performance of our framework using real-world bike trip data and check-in data from New York City and Washington, DC, respectively. Results show that by fusing the two types of urban data, our method achieves fine-grained urban event detection and characterization in both cities and consistently outperforms the baselines.

GreenBicycling: A Smartphone-Based Public Bicycle Sharing System for Healthy Life

Gas-consumed urban transportation causes the problems of environment pollution and traffic congestion. Bicycling is an alternative transportation way for short-distance movement, which is a green and healthy style. Some modern cities are building Public Bicycle Sharing System (PBSS), so that inhabitants and tourists can temporarily use bicycles anywhere in the city. In this paper, Green Bicycling, a smartphone-based Public Bicycle Sharing System for healthy life is developed. Green Bicycling aims to improve the user experience and encourage cyclist to use BSS. It allows cyclist to query not only the current information of rental spots, but also the forecast state. To achieve this goal, an improved back propagation network prediction model is proposed. In addition, a quantitative measurement of calorie consumption in a riding trip is introduced into Green Bicycling, so that the cyclist could get intuitive understanding of how much calorie reduction from the riding. A simplified version of the Green Bicycling APP has been released in the Windows Phone Market.

Understanding bike trip patterns leveraging bike sharing system open data

Bike sharing systems are booming globally as a green and flexible transportationmode, but the flexibility also brings difficulties in keeping the bike stations balanced with enough bikes and docks. Understanding the spatio-temporal bike trip patterns in a bike sharing system, such as the popular trip origins and destinations during rush hours, is important for researchers to design models for bike scheduling and station management. However, due to privacy and operational concerns, bike trip data are usually not publicly available in many cities. Instead, the station feeds about real-time bike and dock number in stations are usually public, which we refer to as bike sharing system open data. In this paper, we propose an approach to infer the spatio-temporal bike trip patterns from the public station feeds. Since the number of possible trips (i.e., origin-destination station pairs) is much larger than the number of stations, we define the trip inference as an ill-posed inverse problem. To solve this problem, we identify the sparsity and locality properties of bike trip patterns, and propose a sparse and weighted regularization model to impose both properties in the solution. We evaluate our method using real-world data fromWashington, D.C. and New York City. Results show that our method can effectively infer the spatio-temporal bike trip patterns and outperform the baselines in both cities.

Inferring bike trip patterns from bike sharing system open data

Understanding bike trip patterns in a bike sharing system is important for researchers designing models for station placement and bike scheduling. By bike trip patterns, we refer to the large number of bike trips observed between two stations. However, due to privacy and operational concerns, bike trip data are usually not made publicly available. In this paper, instead of relying on time-consuming surveys and inaccurate simulations, we attempt to infer bike trip patterns directly from station status data, which are usually public to help riders find nearby stations and bikes. However, the station status data do not contain information about where the bikes come from and go to, therefore the same observations on stations might correspond to different underlying bike trips. To address this challenge, We conduct an empirical study on a sample bike trip dataset to gain insights about the inner structure of bike trips. We then formulate the trip inference problem as an ill-posed inverse problem, and propose a regularization technique to incorporate the a priori information about bike trips to solve the problem. We evaluate our method using real-world bike sharing datasets from Washington, D.C. Results show that our method effectively infers bike trip patterns.

EnUp: Energy-Efficient Data Uploading for Mobile Crowd Sensing Applications

Mobile crowd sensing enables large-scale sensing of the physical world at low cost by leveraging the available sensors on the mobile phones. One of the key factors for the success of mobile crowd sensing is uploading the sensing data to the cloud promptly. Traditional data uploading strategies leveraging whenever available networks may incur extra data cost, impact phone performance,, drain battery power significantly. In this paper, we propose an energy-efficient large data uploading framework using only WiFi network. Specifically, we propose to upload data at WiFi Ready Conditions (WRCs), when the WiFi network is connected, no front-end applications are using it. By forecasting the WRCs that will be encountered in a data uploading task, our framework intelligently selects optimal WRCs to minimize the overall energy consumption. Our evaluation results with the Device Analyzer Dataset show that the proposed method can effectively upload large data while consuming 30% less energy than the greedy-based baseline method.