A Survey on Deep Learning for Human Mobility

The study of human mobility is crucial due to its impact on several aspects of our society, such as disease spreading, urban planning, well-being, pollution, and more. The proliferation of digital mobility data, such as phone records, GPS traces, and social media posts, combined with the predictive power of artificial intelligence, triggered the application of deep learning to human mobility. Existing surveys focus on single tasks, data sources, mechanistic or traditional machine learning approaches, while a comprehensive description of deep learning solutions is missing. This survey provides a taxonomy of mobility tasks, a discussion on the challenges related to each task and how deep learning may overcome the limitations of traditional models, a description of the most relevant solutions to the mobility tasks described above, and the relevant challenges for the future. Our survey is a guide to the leading deep learning solutions to next-location prediction, crowd flow prediction, trajectory generation, and flow generation. At the same time, it helps deep learning scientists and practitioners understand the fundamental concepts and the open challenges of the study of human mobility.

FraudMove: Fraud Drivers Discovery Using Real-Time Trajectory Outlier Detection

Taxicabs and rideshare cars nowadays are equipped with GPS devices that enable capturing a large volume of traces. These GPS traces represent the moving behavior of the car drivers. Indeed, the real-time discovery of fraud drivers earlier is a demand for saving the passenger’s life and money. For this purpose, this paper proposes a novel time-based system, namely FraudMove, to discover fraud drivers in real-time by identifying outlier active trips. Mainly, the proposed FraudMove system computes the time of the most probable path of a trip. For trajectory outlier detection, a trajectory is considered an outlier trajectory if its time exceeds the time of this computed path by a specified threshold. FraudMove employs a tunable time window parameter to control the number of checks for detecting outlier trips. This parameter allows FraudMove to trade responsiveness with efficiency. Unlike other related works that wait until the end of a trip to indicate that it was an outlier, FraudMove discovers outlier trips instantly during the trip. Extensive experiments conducted on a real dataset confirm the efficiency and effectiveness of FraudMove in detecting outlier trajectories. The experimental results prove that FraudMove saves the response time of the outlier check process by up to 65% compared to the state-of-the-art systems.

Land Consumption of Delivery Robots and Bicycle Couriers for On-Demand Meal Delivery Using GPS Data and Simulations Based on the Time-Area Concept

Regulating the curbside usage of delivery vehicles and ride-hailing services as well as micromobility has been a challenge in the last years, a challenge which might worsen with the increase of autonomous vehicles. The contribution of the research outlined in this paper is an evaluation method of the land use of on-demand meal delivery services such as Deliveroo and UberEats. It evaluates the effect parking policies, operating strategy changes, and scheduling options have on the land consumption of bicycle couriers and sidewalk automated delivery robots (SADRs). Various operating strategies (i.e., shared fleets and fleets operated by restaurants), parking policies (i.e., parking at the restaurant, parking at the customer or no parking) and scheduling options (i.e., one meal per vehicle, multiple meals per vehicle) are simulated and applied to New York City (NYC). Additionally, the time-area requirements of on-demand meal delivery services are calculated based on GPS traces of Deliveroo and UberEats riders in two UK cities. The simulation in the paper shows that SADRs can reduce the time-area requirements by half compared with bicycle couriers. The effect of operating strategy changes and forbidding vehicles to park at the customer’s home is small. Delivering multiple meals in one tour halves the time-area requirements. The time-area requirements based on GPS traces is around 300 m2·min per order. The study allows policymakers to learn more about the land use of on-demand meal delivery services and how these can be influenced. Hence, they can adjust their policy strategies to ensure that on-demand meal delivery services are provided in a way that they use land effectively, reduce external costs, improve sustainability and benefit everyone.

Building a Large-Scale Micro-Simulation Transport Scenario Using Big Data

A large-scale agent-based microsimulation scenario including the transport modes car, bus, bicycle, scooter, and pedestrian, is built and validated for the city of Bologna (Italy) during the morning peak hour. Large-scale microsimulations enable the evaluation of city-wide effects of novel and complex transport technologies and services, such as intelligent traffic lights or shared autonomous vehicles. Large-scale microsimulations can be seen as an interdisciplinary project where transport planners and technology developers can work together on the same scenario; big data from OpenStreetMap, traffic surveys, GPS traces, traffic counts and transit details are merged into a unique transport scenario. The employed activity-based demand model is able to simulate and evaluate door-to-door trip times while testing different mobility strategies. Indeed, a utility-based mode choice model is calibrated that matches the official modal split. The scenario is implemented and analyzed with the software SUMOPy/SUMO which is an open source software, available on GitHub. The simulated traffic flows are compared with flows from traffic counters using different indicators. The determination coefficient has been 0.7 for larger roads (width greater than seven meters). The present work shows that it is possible to build realistic microsimulation scenarios for larger urban areas. A higher precision of the results could be achieved by using more coherent data and by merging different data sources.

The effect of GPS refresh rate on measuring police patrol in micro-places

With the increasing prevalence of police interventions implemented in micro hot-spots of crime, the accuracy with which officer foot patrols can be measured is increasingly important for the robust evaluation of such strategies. However, it is currently unknown how the accuracy of GPS traces impact upon our understanding of where officers are at a given time and how this varies for different GPS refresh rates. Most existing studies that use GPS data fail to acknowledge this. This study uses GPS data from police officer radios and ground truth data to estimate how accurate GPS data are for different GPS refresh rates. The similarity of the assumed paths are quantitatively evaluated and the analysis shows that different refresh rates lead to diverging estimations of where officers have patrolled. These results have significant implications for the measurement of police patrols in micro-places and evaluations of micro-place based interventions.

The Fourth ACM SIGSPATIAL Workshop on Location-Based Recommendations, Geosocial Networks and Geoadvertising

The amount of publicly available geo-referenced data has seen a dramatic increase over the last years. Many user activities generate data that are annotated with location and contextual information. Moreover, it has become easier to collect and combine rich and diverse location information. In the context of geoadvertising, the use of geosocial data for targeted marketing is receiving significant attention from a wide spectrum of companies and organizations. With the advent of smartphones and online social networks, a multi-billion dollar industry that utilizes geosocial data for advertising and marketing has emerged. Geotagged social-media posts, GPS traces, data from cellular antennas and WiFi access points are used widely to directly access people for advertising, recommendations, marketing, and group purchases. Exploiting this torrent of geo-referenced data provides a tremendous potential to materially improve existing recommendation services and offer novel ones, with numerous applications in many domains, including social networks, marketing, and tourism.