[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-121260-en":3,"doc-seo-121260-105":30,"detail-sidebar-cat-0-en-105":92},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},121260,1374391975076,"Riley","https://ap-avatar.wpscdn.com/avatar/14000253ca4ec9f6853?x-image-process=image/resize,m_fixed,w_180,h_180&k=1783305029341752051",8,"Research & Report","Using GPS Data to Predict Accident Risk - A Data Mining and Machine Learning Approach","Rapid growth of digital data and improved computation enable new approaches to accident analysis and broader social science research. Traditional accident-analysis practice often lags in adopting machine learning and data-mining methods and may confuse explanatory power with predictive power. This paper establishes theoretical foundations for these methods in the digital era and demonstrates their application to a GPS dataset and an insurance dataset to enhance accident prediction modeling.","THE UNIVERSITY OF CHICAGO  \nUsing GPS Data to Predict Accident Risk: A Data Mining and Machine Learning Approach  \nBy  \nKanyao Han  \nJuly 2018  \nA paper submitted in partial fulfillment of the requirements for the  \nMaster of Arts degree in the Master of Arts Program in the Social Sciences  \nFaculty Advisor: Muh-Chung Lin  \nPreceptor: Caterina Fugazzola  \nAbstract  \nThe rapid accumulation of digital data, coupled with advances in computational capacity, has ushered in a new era of data exploration and analysis. This shift presents significant opportunities for both accident analysis and broader social science research. Despite these developments, traditional social science disciplines, including accident analysis, have been slow to adopt powerful machine learning and data mining techniques. Moreover, they often conflate explanatory power with predictive power. This paper aims to (1) examine the theoretical foundations of machine learning and data mining within the context of the digital era, and (2) demonstrate the application of these methods to a GPS dataset and an insurance dataset for the purpose of enhancing accident prediction modeling.  \n1. Introduction  \nThe popularization of digital technology over the last decade has ushered in a new era for both data science and social science. Current digital data, which directly or indirectly embodies individual-level human behaviors and socioeconomic information, are accumulated at an unprecedented speed. It thus brings many new opportunities to social science research. Among various types of data, Global Position System (GPS) data has proven to be one of the most promising types, especially in accident analysis and prediction for public policy and insurance strategy (Karapiperis et al. 2015) . This is because we can directly obtain individual-level driving behavioral information from it, which are usually not included in conventional insurance surveys (Jin, Deng, and Jiang 2018) . Besides, compared to survey data, digital observational data also has two distinctive advantages: always-on and non-reactive (Salganik 2018) . For the first advantage, a telematics device can constantly collect data as long as a car is running. As to the second advantage, the device must provide unbiased records, while the information documented in insurance surveys often has a bias due to drivers’ reaction to possible insurance plans. For example, White (1976) finds that the self-reported driving mileage in insurance surveys is usually lower than the actual mileage since drivers know that higher self-reported mileage can lead to a more expansive plan.  \nBecause of the potential of GPS data mentioned above and the recent commercialization of the concept of Usage-Based Insurance (UBI), also known as Pay-As-You-Drive (PAYD), in the car insurance industry (Karapiperis et al. 2015), there have been several studies that try to extract behavioral features from GPS data and analyze their roles in  \naccident risk. For example, actual driving mileage (Paefgen, Staake, and Fleisch 2014, Lemaire, Park, and Wang 2016, Litman 2005, Elvik 2015) , rates of hard accelerations or hard brakes (Weidner, Transchel, and Weidner 2017, Handel et al. 2014, Bagdadi and Várhelyi 2011, Paefgen, Staake, and Fleisch 2014) , strategic driving behaviors such as road and time selection (Tselentis, Yannis, and Vlahogianni 2017), and daily driving behaviorsand mobility patterns such as nighttime driving and familiarity with driving routes (Jin, Deng, and Jiang 2018, Ayuso, Guillén, and Pérez-Marín 2014, 2016, Paleti, Eluru, and Bhat 2010, Behnood, Roshandeh, and Mannering 2014, Eluru et al. 2012) have been used to build statistical models for accident analysis and prediction in specific cities.  \nHowever, although the vast majority of previous studies directly or indirectly claim that their ultimate goal is for prediction, the models they use are typically explanatory, instead of predictive (it will be discussed later) . This phenomeno","cbCaif4kzj7Bo3vp","https://ap.wps.com/l/cbCaif4kzj7Bo3vp","pdf",1166501,1,43,"English","en",105,"# Introduction\n## GPS data and digital observational advantages\n## Predictive vs explanatory models\n# Two Cultures in Statistics\n## Two cultures","[{\"question\":\"What problem does the paper address about existing accident prediction research?\",\"answer\":\"It highlights that many studies claim predictive goals but rely on explanatory models, and that explanatory power is often mistaken for predictive power.\"},{\"question\":\"Why is GPS data considered promising for accident analysis?\",\"answer\":\"GPS data can capture individual-level driving behavior directly and, unlike surveys, offers always-on and non-reactive observations suitable for modeling and policy-relevant analysis.\"},{\"question\":\"What does the paper aim to do with machine learning and data mining?\",\"answer\":\"It introduces predictive modeling foundations and applies machine learning methods to GPS and insurance datasets to improve accident prediction modeling.\"}]","Using GPS Data to Predict Accident Risk - A Data Mining and Machine Learning Approach | PDF",1785734735,108,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":28},"using-gps-data-to-predict-accident-risk-a-data-mining-and-machine-learning-approach","",{"@graph":36,"@context":86},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/using-gps-data-to-predict-accident-risk-a-data-mining-and-machine-learning-approach/121260/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05","2026-08-03",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What problem does the paper address about existing accident prediction research?","Question",{"text":76,"@type":77},"It highlights that many studies claim predictive goals but rely on explanatory models, and that explanatory power is often mistaken for predictive power.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"Why is GPS data considered promising for accident analysis?",{"text":81,"@type":77},"GPS data can capture individual-level driving behavior directly and, unlike surveys, offers always-on and non-reactive observations suitable for modeling and policy-relevant analysis.",{"name":83,"@type":74,"acceptedAnswer":84},"What does the paper aim to do with machine learning and data mining?",{"text":85,"@type":77},"It introduces predictive modeling foundations and applies machine learning methods to GPS and insurance datasets to improve accident prediction modeling.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]