Spatio-Temporal Graph Attention Embedding for Joint Crowd Flow and Transition Predictions: A Wi-Fi-based Mobility Case Study

Abstract

Crowd mobility prediction, in particular, forecasting flows at and transitions across different locations, is essential for crowd analytics and management in spacious environments featured with large gathering. We propose GAEFT, a novel crowd mobility analytics system based on the multi-task graph attention neural network to forecast crowd flows (inflows/outflows) and transitions. Specifically, we leverage the collective and sanitized campus Wi-Fi association data provided by our university information technology service and conduct a relatable case study. Our comprehensive data analysis reveals the important challenges of sparsity and skewness, as well as the complex spatio-temporal variations within the crowd mobility data. Therefore, we design a novel spatio-temporal clustering method to group Wi-Fi access points (APs) with similar transition features, and obtain more regular mobility features for model inputs. We then propose an attention-based graph embedding design to capture the correlations among the crowd flows and transitions, and jointly predict the AP-level flows as well as transitions across buildings and clusters through a multi-task formulation. Extensive experimental studies using more than 28 million association records collected during 2020-2021 academic year validate the excellent accuracy of GAEFT in forecasting dynamic and complex crowd mobility.

References

1. Serdar Çolak, Antonio Lima, and Marta C González. 2016. Understanding congested travel in urban areas. Nature Communications 7, 1 (2016), 1--8.
2. Allan M De Souza, Roberto S Yokoyama, Guilherme Maia, Antonio Loureiro, and Leandro Villas. 2016. Real-time path planning to prevent traffic jam through an intelligent transportation system. In Proc. IEEE ISCC. IEEE, 726--731.
3. Zipei Fan, Xuan Song, Tianqi Xia, Renhe Jiang, Ryosuke Shibasaki, and Ritsu Sakuramachi. 2018. Online deep ensemble learning for predicting citywide human mobility. Proc. ACM IMWUT 2, 3 (2018), 1--21.
4. Zhihan Fang, Yu Yang, Shuai Wang, Boyang Fu, Zixing Song, Fan Zhang, and Desheng Zhang. 2019. MAC: Measuring the impacts of anomalies on travel time of multiple transportation systems. Proc. ACM IMWUT 3, 2 (2019), 1--24.
5. Jie Feng, Can Rong, Funing Sun, Diansheng Guo, and Yong Li. 2020. PMF: A privacy-preserving human mobility prediction framework via federated learning. Proc. ACM IMWUT 4, 1 (2020), 1--21.
6. Brendan J Frey and Delbert Dueck. 2007. Clustering by passing messages between data points. Science 315, 5814 (2007), 972--976.
7. Chen Gao, Chao Huang, Yue Yu, Huandong Wang, Yong Li, and Depeng Jin. 2019. Privacy-preserving cross-domain location recommendation. Proc. ACM IMWUT 3, 1 (2019), 1--21.
8. Xu Geng, Yaguang Li, Leye Wang, Lingyu Zhang, Qiang Yang, Jieping Ye, and Yan Liu. 2019. Spatiotemporal multi-graph convolution network for ride-hailing demand forecasting. In Proc. AAAI, Vol. 33. 3656--3663.
9. Craig Gentry and Zulfikar Ramzan. 2005. Single-database private information retrieval with constant communication rate. In International Colloquium on Automata, Languages, and Programming. Springer, 803--815.
10. Yongshun Gong, Zhibin Li, Jian Zhang, Wei Liu, and Yu Zheng. 2020. Online spatio-temporal crowd flow distribution prediction for complex metro system. IEEE TKDE (2020).
11. Anhong Guo, Anuraag Jain, Shomiron Ghose, Gierad Laput, Chris Harrison, and Jeffrey P Bigham. 2018. Crowd-AI camera sensing in the real world. Proc. ACM IMWUT 2, 3 (2018), 1--20.
12. Suining He and S.-H. Gary Chan. 2015. Wi-Fi fingerprint-based indoor positioning: Recent advances and comparisons. IEEE Communications Surveys & Tutorials 18, 1 (2015), 466--490.
13. Suining He, Tianyang Hu, and S.-H. Gary Chan. 2015. Contour-based trilateration for indoor fingerprinting localization. In Proc. ACM SenSys. 225--238.
14. Suining He and Kang G. Shin. 2018. Steering Crowdsourced Signal Map Construction via Bayesian Compressive Sensing. In Proc. IEEE INFOCOM. 1016--1024.
15. Suining He and Kang G. Shin. 2019. Crowd-flow graph construction and identification with spatio-temporal signal feature fusion. In Proc. IEEE INFOCOM. 757--765.
16. Suining He and Kang G. Shin. 2019. Spatio-Temporal Capsule-Based Reinforcement Learning for Mobility-on-Demand Network Coordination. In Proc. WWW (San Francisco, CA, USA). 2806--2813.
17. Suining He and Kang G. Shin. 2020. Dynamic Flow Distribution Prediction for Urban Dockless E-Scooter Sharing Reconfiguration. In Proc. WWW. 133--143.
18. Suining He and Kang G. Shin. 2020. Towards fine-grained flow forecasting: A graph attention approach for bike sharing systems. In Proc. WWW. 88--98.
19. Jilin Hu, Bin Yang, Chenjuan Guo, Christian S Jensen, and Hui Xiong. 2020. Stochastic origin-destination matrix forecasting using dual-stage graph convolutional, recurrent neural networks. In Proc. IEEE ICDE. 1417--1428.
20. Tiziano Inzerilli, Anna Maria Vegni, Alessandro Neri, and Roberto Cusani. 2008. A location-based vertical handover algorithm for limitation of the ping-pong effect. In Proc. IEEE WiMOB. IEEE, 385--389.
21. Renhe Jiang, Xuan Song, Dou Huang, Xiaoya Song, Tianqi Xia, Zekun Cai, Zhaonan Wang, Kyoung-Sook Kim, and Ryosuke Shibasaki. 2019. DeepUrbanEvent: A system for predicting citywide crowd dynamics at big events. In Proc. ACM SIGKDD. 2114--2122.
22. Yaguang Li, Rose Yu, Cyrus Shahabi, and Yan Liu. 2017. Diffusion convolutional recurrent neural network: Data-driven traffic forecasting. arXiv preprint arXiv:1707.01926 (2017).
23. Yexin Li, Yu Zheng, Huichu Zhang, and Lei Chen. 2015. Traffic prediction in a bike-sharing system. In Proc. ACM SIGSPATIAL. 1--10.
24. Lei Lin, Zhengbing He, and Srinivas Peeta. 2018. Predicting station-level hourly demand in a large-scale bike-sharing network: A graph convolutional neural network approach. Transportation Research Part C: Emerging Technologies 97 (2018), 258--276.
25. Ziqian Lin, Jie Feng, Ziyang Lu, Yong Li, and Depeng Jin. 2019. DeepSTN+: Context-aware spatial-temporal neural network for crowd flow prediction in metropolis. In Proc. AAAI, Vol. 33. 1020--1027.
26. Lingbo Liu, Jingwen Chen, Hefeng Wu, Jiajie Zhen, Guanbin Li, and Liang Lin. 2020. Physical-virtual collaboration modeling for intra-and inter-station metro ridership prediction. IEEE TITS (2020).
27. Lingbo Liu, Zhilin Qiu, Guanbin Li, Qing Wang, Wanli Ouyang, and Liang Lin. 2019. Contextualized spatial-temporal network for taxi origin-destination demand prediction. IEEE TITS 20, 10 (2019), 3875--3887.
28. Qian Liu, Dexuan Sha, Wei Liu, Paul Houser, Luyao Zhang, Ruizhi Hou, Hai Lan, Colin Flynn, Mingyue Lu, Tao Hu, et al. 2020. Spatiotemporal patterns of COVID-19 impact on human activities and environment in mainland China using nighttime light and air quality data. Remote Sensing 12, 10 (2020), 1576.
29. Zhicheng Liu, Fabio Miranda, Weiting Xiong, Junyan Yang, Qiao Wang, and Claudio Silva. 2020. Learning geo-contextual embeddings for commuting flow prediction. In Proc. AAAI, Vol. 34. 808--816.
30. Jamie Lopez Bernal, Nick Andrews, Charlotte Gower, Eileen Gallagher, Ruth Simmons, Simon Thelwall, Julia Stowe, Elise Tessier, Natalie Groves, Gavin Dabrera, et al. 2021. Effectiveness of COVID-19 vaccines against the B. 1.617. 2 (Delta) variant. New England Journal of Medicine (2021).
31. Zheyi Pan, Yuxuan Liang, Weifeng Wang, Yong Yu, Yu Zheng, and Junbo Zhang. 2019. Urban traffic prediction from spatio-temporal data using deep meta learning. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 1720--1730.
32. Piotr Sapiezynski, Arkadiusz Stopczynski, David Kofoed Wind, Jure Leskovec, and Sune Lehmann. 2017. Inferring person-to-person proximity using WiFi signals. Proc. ACM IMWUT 1, 2 (2017), 1--20.
33. Jiaxing Shen, Jiannong Cao, and Xuefeng Liu. 2019. BaG: Behavior-aware Group Detection in Crowded Urban Spaces using WiFi Probes. In Proc. WWW. 1669--1678.
34. Shun-Yao Shih, Fan-Keng Sun, and Hung-yi Lee. 2019. Temporal pattern attention for multivariate time series forecasting. Machine Learning 108, 8 (2019), 1421--1441.
35. Yiwei Song, Yunhuai Liu, Wenqing Qiu, Zhou Qin, Chang Tan, Can Yang, and Desheng Zhang. 2020. MIFF: Human Mobility Extractions with Cellular Signaling Data under Spatio-temporal Uncertainty. Proc. ACM IMWUT 4, 4 (2020), 1--19.
36. Waldo R Tobler. 1970. A computer movie simulating urban growth in the Detroit region. Economic Geography 46, sup1 (1970), 234--240.
37. Florian Toqué, Etienne Côme, Mohamed Khalil El Mahrsi, and Latifa Oukhellou. 2016. Forecasting dynamic public transport origin-destination matrices with long-short term memory recurrent neural networks. In Proc. IEEE ITSC. 1071--1076.
38. Martin W Traunmueller, Nicholas Johnson, Awais Malik, and Constantine E Kontokosta. 2018. Digital footprints: Using WiFi probe and locational data to analyze human mobility trajectories in cities. Computers, Environment and Urban Systems 72 (2018), 4--12.
39. Amee Trivedi, Camellia Zakaria, Rajesh Balan, Ann Becker, George Corey, and Prashant Shenoy. 2021. WiFiTrace: Network-based Contact Tracing for Infectious Diseases Using Passive WiFi Sensing. Proc. ACM IMWUT 5, 1 (2021), 1--26.
40. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need. In Proc. NeurIPS. 5998--6008.
41. Petar Velišković, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Lio, and Yoshua Bengio. 2017. Graph attention networks. arXiv preprint arXiv:1710.10903 (2017).
42. Yuandong Wang, Hongzhi Yin, Hongxu Chen, Tianyu Wo, Jie Xu, and Kai Zheng. 2019. Origin-destination matrix prediction via graph convolution: a new perspective of passenger demand modeling. In Proc. ACM SIGKDD. 1227--1235.
43. Xi Xiong, Kaan Ozbay, Li Jin, and Chen Feng. 2020. Dynamic origin-destination matrix prediction with line graph neural networks and Kalman filter. Transportation Research Record 2674, 8 (2020), 491--503.
44. Xi Yang, Suining He, and Huiqun Huang. 2020. Station Correlation Attention Learning for Data-driven Bike Sharing System Usage Prediction. In Proc. IEEE MASS. 640--648.
45. Zijun Yao, Yanjie Fu, Bin Liu, Yanchi Liu, and Hui Xiong. 2016. POI recommendation: A temporal matching between POI popularity and user regularity. In Proc. IEEE ICDM. 549--558.
46. Nicholas Jing Yuan, Yu Zheng, Xing Xie, Yingzi Wang, Kai Zheng, and Hui Xiong. 2014. Discovering urban functional zones using latent activity trajectories. IEEE TKDE 27, 3 (2014), 712--725.
47. Huichu Zhang, Yu Zheng, and Yong Yu. 2018. Detecting urban anomalies using multiple spatio-temporal data sources. Proc. ACM IMWUT 2, 1 (2018), 1--18.
48. Junbo Zhang, Yu Zheng, and Dekang Qi. 2017. Deep spatio-temporal residual networks for citywide crowd flows prediction. In Proc. AAAI, Vol. 31.
49. Chuanpan Zheng, Xiaoliang Fan, Cheng Wang, and Jianzhong Qi. 2020. GMAN: A graph multi-attention network for traffic prediction. In Proc. AAAI, Vol. 34. 1234--1241.
50. Yuren Zhou, Billy Pik Lik Lau, Zann Koh, Chau Yuen, and Benny Kai Kiat Ng. 2020. Understanding crowd behaviors in a social event by passive WiFi sensing and data mining. IEEE IoT-J 7, 5 (2020), 4442--4454.

Index Terms

• Information systems

  • Information systems applications

    • Mobile information processing systems