{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/bike-flow-prediction-with-multi-graph","title":"Bike Flow Prediction with Multi-Graph Convolutional Networks","arxiv_id":"1807.10934","date":"2018-07-28","proceeding":null,"authors":["Di Chai","Leye Wang","Qiang Yang"],"abstract":"One fundamental issue in managing bike sharing systems is the bike flow\nprediction. Due to the hardness of predicting the flow for a single station,\nrecent research works often predict the bike flow at cluster-level. While such\nstudies gain satisfactory prediction accuracy, they cannot directly guide some\nfine-grained bike sharing system management issues at station-level. In this\npaper, we revisit the problem of the station-level bike flow prediction, aiming\nto boost the prediction accuracy leveraging the breakthroughs of deep learning\ntechniques. We propose a new multi-graph convolutional neural network model to\npredict the bike flow at station-level, where the key novelty is viewing the\nbike sharing system from the graph perspective. More specifically, we construct\nmultiple inter-station graphs for a bike sharing system. In each graph, nodes\nare stations, and edges are a certain type of relations between stations. Then,\nmultiple graphs are constructed to reflect heterogeneous relationships (e.g.,\ndistance, ride record correlation). Afterward, we fuse the multiple graphs and\nthen apply the convolutional layers on the fused graph to predict station-level\nfuture bike flow. In addition to the estimated bike flow value, our model also\ngives the prediction confidence interval so as to help the bike sharing system\nmanagers make decisions. Using New York City and Chicago bike sharing data for\nexperiments, our model can outperform state-of-the-art station-level prediction\nmodels by reducing 25.1% and 17.0% of prediction error in New York City and\nChicago, respectively.","url_abs":"http://arxiv.org/abs/1807.10934v1","url_pdf":"http://arxiv.org/pdf/1807.10934v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"bike-flow-prediction-with-multi-graph","repo_url":"https://github.com/Di-Chai/GraphCNN-Bike","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"management","task_name":"Management"},{"task_slug":"prediction","task_name":"Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1807.10934","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}