Graph Neural Networks for Road Safety Modeling: Datasets and Evaluations for Accident Analysisdoi:10.7910/DVN/V71K5RHarvard Dataverse2023-10-262Abhinav Nippani; Dongyue Li, 2023, "Graph Neural Networks for Road Safety Modeling: Datasets and Evaluations for Accident Analysis", https://doi.org/10.7910/DVN/V71K5R, Harvard Dataverse, V2Graph Neural Networks for Road Safety Modeling: Datasets and Evaluations for Accident AnalysisAccompanying Datasetsdoi:10.7910/DVN/V71K5RAbhinav NippaniDongyue LiHarvard DataverseRyan ZhangComputer and Information ScienceGraph Neural Networks, Traffic Accident Analysis, Road NetworksHere we deposit the datasets we have extracted for ten states in the US. In each zip file, we include each state's accident records, road networks, and network features. For further information about using the dataset and how we extracted the data, check out our GitHub repository for instructions.For each state, because there are multiple files in the dataset, we uploaded a zip file that includes all the files for that state. As a summary, once you download the zip file, unzip it, then there should be the following files in it: (1) adj_matrix.pt: The sparse adjacency matrix of the road network. (2) accidents_monthly.csv: All accidents spanning multiple years and aggregated by month. (3) Nodes/: The node features, including weather information, every month. (4) node_features_{year}_{month}.pt: The weather information of a particular month. (5) Edges/: The edge features, including road and traffic volume information, if available. (6) edge_features.pt: Edge features describing the road information. (7) edge_features_traffic_{year}.pt: Traffic volume records of a particular year.<a href="http://creativecommons.org/publicdomain/zero/1.0">CC0 1.0</a>CA-1.zipapplication/zipCA-2.zipapplication/zipDE.zipapplication/zipIA.zipapplication/zipIL.zipapplication/zipMA.zipapplication/zipMD.zipapplication/zipMN.zipapplication/zipMT.zipapplication/zipNV.zipapplication/zipNY.zipapplication/zip