{"title":"海报:安全路线:评估道路安全的框架","authors":"Reuben Vince Rabsatt, H. Kalantarian, M. Gerla","doi":"10.1109/VNC.2016.7835957","DOIUrl":null,"url":null,"abstract":"Navigation systems typically suggest directions to the user based on factors such as weather conditions, traffic information, and road hazards. By considering these factors, the system can suggest a route in which travel time is minimized. However, these navigation systems fail to include any meaningful information about the relative safety of different routes. For example, some roads are significantly more accident prone than others and can easily be avoided. By taking alternative paths, individuals can reduce risk, while often adding only minimal time to their commutes. We develop a model for road safety based on data from the California Freeway Performance Measurement System (PEMS) database. Our model for road safety considers risk using historical accident data based on factors such as the time of day, day of week, average speed, flow, and other traffic features. Our experimental results using the RandomForest classifier show a classification accuracy of 80% for identifying high-risk routes.","PeriodicalId":352428,"journal":{"name":"2016 IEEE Vehicular Networking Conference (VNC)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Poster: SafeRoute a framework for assessment of road safety\",\"authors\":\"Reuben Vince Rabsatt, H. Kalantarian, M. Gerla\",\"doi\":\"10.1109/VNC.2016.7835957\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Navigation systems typically suggest directions to the user based on factors such as weather conditions, traffic information, and road hazards. By considering these factors, the system can suggest a route in which travel time is minimized. However, these navigation systems fail to include any meaningful information about the relative safety of different routes. For example, some roads are significantly more accident prone than others and can easily be avoided. By taking alternative paths, individuals can reduce risk, while often adding only minimal time to their commutes. We develop a model for road safety based on data from the California Freeway Performance Measurement System (PEMS) database. Our model for road safety considers risk using historical accident data based on factors such as the time of day, day of week, average speed, flow, and other traffic features. Our experimental results using the RandomForest classifier show a classification accuracy of 80% for identifying high-risk routes.\",\"PeriodicalId\":352428,\"journal\":{\"name\":\"2016 IEEE Vehicular Networking Conference (VNC)\",\"volume\":\"12 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2016-12-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2016 IEEE Vehicular Networking Conference (VNC)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/VNC.2016.7835957\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2016 IEEE Vehicular Networking Conference (VNC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/VNC.2016.7835957","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Poster: SafeRoute a framework for assessment of road safety
Navigation systems typically suggest directions to the user based on factors such as weather conditions, traffic information, and road hazards. By considering these factors, the system can suggest a route in which travel time is minimized. However, these navigation systems fail to include any meaningful information about the relative safety of different routes. For example, some roads are significantly more accident prone than others and can easily be avoided. By taking alternative paths, individuals can reduce risk, while often adding only minimal time to their commutes. We develop a model for road safety based on data from the California Freeway Performance Measurement System (PEMS) database. Our model for road safety considers risk using historical accident data based on factors such as the time of day, day of week, average speed, flow, and other traffic features. Our experimental results using the RandomForest classifier show a classification accuracy of 80% for identifying high-risk routes.