Towards Dynamic Crowd Mobility Learning and Meta Model Updates for A Smart Connected Campus
Abstract
In this paper, we propose MetaMobi, a novel spatiotemporal multi-dots connectivity-aware modeling and Meta model update approach for crowd Mobility learning. MetaMobi analyzes real-world Wi-Fi association data collected from our campus wireless infrastructure, with the goal towards enabling a smart connected campus. Specifically, MetaMobi aims at addressing the following two major challenges with existing crowd mobility sensing system designs: (a) how to handle the spatially, temporally, and contextually varying features in large-scale human crowd mobility distributions; and (b) how to adapt to the impacts of such crowd mobility patterns as well as the dynamic changes in crowd sensing infrastructures. To handle the first challenge, we design a novel multi-dots connectivity-aware learning approach, which jointly learns the crowd flow time series of multiple buildings with fusion of spatial graph connectivities and temporal attention mechanisms. Furthermore, to overcome the adaptivity issues due to changes in the crowd sensing infrastructures (e.g., installation of new access points), we further design a novel meta model update approach with Bernoulli dropout, which mitigates the overfitting behaviors of the model given few-shot distributions of new crowd mobility datasets. Extensive experimental evaluations based on the real-world campus wireless dataset (including over 76 million Wi-Fi association and disassociation records) demonstrate the accuracy, effectiveness, and adaptivity of MetaMobi in forecasting the campus crowd flows, with 30% higher accuracy compared to the state-of-the-art approaches.
References
[1]
Breiman,L. 2001. "Random forests. Machine learning". In Online-ArXiV Preprint or similar. vol. 45,pp. 5--32.
[2]
Kipf,T N.,Welling,M. 2016. "Semi-supervised classification with graph convolutional networks". In Online-ArXiV Preprint or similar.
[3]
Froehlich,J E.,Neumann,J., and Oliver,N. 2009. "Sensing and predicting the pulse of the city through shared bicycling". In Proc. IJCAI.
[4]
Chu,J.,Wang,X.,Qian,K.,Yao,L.,Xiao,F.,Li,J., and Yang,Z. 2020. "Passenger demand prediction with cellular footprints". In IEEE TMC.
[5]
Lin,Z.,Feng,J.,Lu,Z.,Li,Y., and Jin,D. 2019. "DeepSTN+: Context-aware spatial-temporal neural network for crowd flow prediction in metropolis". In Proc. AAAI. vol. 33,pp. 1020--1027.
[6]
Nichol,A.,Schulman,J. 2018. "Reptile: a scalable metalearning algorithm". In Online-ArXiV Preprint or similar. vol. 2,pp. 4--4.
[7]
Lim,K H.,Chan,J.,Karunasekera,S., and Leckie,C. 2019. "Tour recommendation and trip planning using location-based social media: A survey". In Knowledge and Information Systems. vol. 60,pp. 1247--1275.
[8]
Veličkovi´veličkovi´c,P.,Cucurull,G.,Casanova,A.,Romero,A.,Lio,P., and Bengio,Y. 2017. "Graph attention networks". In Online-ArXiV Preprint or similar.
[9]
Chen,L.,Jakubowicz,J.,Yang,D.,Zhang,D., and Pan,G. 2016. "Fine-grained urban event detection and characterization based on tensor cofactorization". In IEEE Transactions on Human-Machine Systems. vol. 47,pp. 380--391.
[10]
Zhang,J.,Zheng,Y.,Qi,D.,Li,R.,Yi,X., and Li,T. 2018. "Predicting citywide crowd flows using deep spatio-temporal residual networks". In Artificial Intelligence. vol. 259,pp. 147--166.
[11]
Jiang,R.,Cai,Z.,Wang,Z.,Yang,C.,Fan,Z.,Chen,Q.,Song,X., and Shibasaki,R. 2022. "Predicting citywide crowd dynamics at big events: A deep learning system". In ACM TIST. vol. 13,pp. 1--24.
[12]
Wang,L.,Chai,D.,Liu,X.,Chen,L., and Chen,K. 2021. "Exploring the generalizability of spatio-temporal traffic prediction: Meta-modeling and an analytic framework". In IEEE TKDE
[13]
Abdellah,A R.,Koucheryavy,A. 2020. "Deep learning with long shortterm memory for iot traffic prediction". In Internet of Things, Smart Spaces, and Next Generation Networks and Systems. pp. 267--280.
[14]
Gentry,C.,Ramzan,Z. 2005. "Single-database private information retrieval with constant communication rate". In Proc. ICALP. pp. 803--815.
[15]
Finn,C.,Abbeel,P., and Levine,S. 2017. "Model-agnostic meta-learning for fast adaptation of deep networks". In ICML. pp. 1126--1135.
[16]
Antoniou,A.,Edwards,H., and Storkey,A. 2018. "How to train your MAML". In Online-ArXiV Preprint or similar.
[17]
Lin,L.,He,Z., and Peeta,S. 2018. "Predicting station-level hourly demand in a large-scale bike-sharing network: A graph convolutional neural network approach". In Transportation Research Part C: Emerging Technologies. vol. 97,pp. 258--276.
[18]
Quinlan,J R. 1986. "Induction of decision trees". In Machine Learning. vol. 1,pp. 81--106.
[19]
He,K.,Zhang,X.,Ren,S., and Sun,J. 2016. "Deep residual learning for image recognition". In Proc. IEEE CVPR. pp. 770--778.
[20]
Shih,S.-Y.,Sun,F.-K., and Lee,H.-Y. 2019. "Temporal pattern attention for multivariate time series forecasting". In Machine Learning. vol. 108,pp. 1421--1441.
[21]
Kingma,D P.,Ba,J. 2014. "Adam: A method for stochastic optimization". In Online-ArXiV Preprint or similar.
[22]
Hochreiter,S.,Schmidhuber,J. 1997. "Long short-term memory". In Neural computation. vol. 9,pp. 1735--1780.
[23]
Liu,M.,Li,L.,Li,Q.,Bai,Y., and Hu,C. 2021. "Pedestrian flow prediction in open public places using graph convolutional network". In ISPRS International Journal of Geo-Information. vol. 10,pp. 455--455.
[24]
Zhang,J.,Zheng,Y., and Qi,D. 2017. "Deep spatio-temporal residual networks for citywide crowd flows prediction". In Proc. AAAI. vol. 31,
[25]
Sung,F.,Yang,Y.,Zhang,L.,Xiang,T.,Torr,P H., and Hospedales,T M. 2018. "Learning to compare: Relation network for few-shot learning". In Proc. IEEE CVPR. pp. 1199--1208.
[26]
Jiang,R.,Cai,Z.,Wang,Z.,Yang,C.,Fan,Z.,Chen,Q.,Tsubouchi,K.,Song,X., and Shibasaki,R. 2021. "DeepCrowd: A deep model for large-scale citywide crowd density and flow prediction". In IEEE TKDE.
[27]
Fang,Z.,Fu,B.,Qin,Z.,Zhang,F., and Zhang,D. 2020. "PrivateBus: Privacy Identification and Protection in Large-Scale Bus WiFi Systems". In Proc. ACM IMWUT. vol. 4,
[28]
Wang,L.,Geng,X.,Ma,X.,Liu,F., and Yang,Q. 2018. "Crowd flow prediction by deep spatio-temporal transfer learning". In Online-ArXiV Preprint or similar.
[29]
Wang,X.,Zhou,Z.,Zhao,Y.,Zhang,X.,Xing,K.,Xiao,F.,Yang,Z., and Liu,Y. 2019. "Improving urban crowd flow prediction on flexible region partition". In IEEE TMC. vol. 19,pp. 2804--2817.
[30]
Tian,P.,Li,W., and Gao,Y. 2021. "Consistent meta-regularization for better meta-knowledge in few-shot learning." In IEEE TNNLS
[31]
Pan,Z.,Zhang,W.,Liang,Y.,Zhang,W.,Yu,Y.,Zhang,J., and Zheng,Y. 2020. "Spatio-temporal meta learning for urban traffic prediction". In IEEE TKDE.
[32]
Wu,Z.,Pan,S.,Long,G.,Jiang,J., and Zhang,C. 2019. "Graph wavenet for deep spatial-temporal graph modeling". In Online-ArXiV Preprint or similar.
[33]
Rajendran,J.,Irpan,A., and Jang,E. 2020. "Meta-learning requires metaaugmentation". In Proc. NeurIPS. vol. 33,pp. 5705--5715.
[34]
Draghici,A.,Agiali,T., and Chilipirea,C. 2015. "Visualization system for human mobility analysis". In Proc. RoEduNet NER. pp. 152--157.
[35]
Yao,H.,Liu,Y.,Wei,Y.,Tang,X., and Li,Z. 2019. "Learning from multiple cities: A meta-learning approach for spatial-temporal prediction". In Proc. WWW. pp. 2181--2191.
[36]
Zhao,Y.,Wang,X.,Li,J.,Zhang,D., and Yang,Z. 2019. "CellTrans: Private Car or Public Transportation? Infer Users' Main Transportation Modes at Urban Scale with Cellular Data". In Proc. ACM IMWUT. vol. 3,
[37]
Zeng,D.,Cao,Z., and Neill,D B. 2021. "Artificial intelligence-enabled public health surveillance-from local detection to global epidemic monitoring and control". In Artificial Intelligence in Medicine. pp. 437--453.
[38]
Hu,X.,Zheng,H.,Chen,Y., and Chen,L. 2015. "Dense crowd counting based on perspective weight model using a fisheye camera". In Optik. vol. 126,pp. 123--130.
[39]
Tseng,H.-Y.,Chen,Y.-W.,Tsai,Y.-H.,Liu,S.,Lin,Y.-Y., and Yang,M.-H. 2020. "Regularizing meta-learning via gradient dropout". In Proc. ACM ACCV.
[40]
Yang,X.,He,S.,Wang,B., and Tabatabaie,M. 2022. "Spatio-Temporal Graph Attention Embedding for Joint Crowd Flow and Transition Predictions: A Wi-Fi-Based Mobility Case Study". In Proc. ACM IMWUT. vol. 5,
[41]
Tabatabaie,M.,He,S., and Yang,X. 2022. "Driver maneuver identification with multi-representation learning and meta model update designs". In Proc. ACM IMWUT. vol. 6,
Index Terms
- Towards Dynamic Crowd Mobility Learning and Meta Model Updates for A Smart Connected Campus
Index terms have been assigned to the content through auto-classification.
Recommendations
Animating the Crowd Mirage: A WiFi-Positioning-Based Crowd Mobility Digital Twin for Smart Campuses
Understanding crowd mobility is critical for many applications. In this paper, we propose CrowdMirage, a WiFi positioning-based crowd mobility digital twin for smart campuses. Specifically, we first design an end-to-end human mobility trace extraction ...
Smart mobility management in 5G heterogeneous networks
Mobility management has always been a challenging and important issue for the next‐generation all‐IP mobile networks. Seamless mobility is one of the requirements to provide real‐time services in such networks. Proxy Mobile IPv6 (PMIPv6) has been ...
Mining crowd mobility and WiFi hotspots on a densely-populated campus
UbiComp '17: Proceedings of the 2017 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2017 ACM International Symposium on Wearable ComputersUnderstanding crowd activities at large-scale and diagnosing existing problems of planning on densely-populated campus are fundamentally hard through traditional ways of measurement and management. In this paper, we demonstrate how to collect data from ...
Comments
Information & Contributors
Information
Published In
December 2022
273 pages
Sponsors
In-Cooperation
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 18 January 2023
Check for updates
Author Tags
Qualifiers
- Article
Conference
EWSN '22
Sponsor:
Acceptance Rates
EWSN '22 Paper Acceptance Rate 18 of 46 submissions, 39%;
Overall Acceptance Rate 81 of 195 submissions, 42%
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 09 Jan 2025