The map shows predicted cycling safety levels for each street segment in DC. Clicking on the segment shows information about the street and the certainty of the safety label. These values are computed using a set of machine learning methods that predict cycling safety based on built environment and social characteristics of the street. More details in the paper:

Wu, Jiahui, Lingzi Hong, and Vanessa Frias-Martinez. "Predicting Perceived Cycling Safety Levels Using Open and Crowdsourced Data." 2018 IEEE International Conference on Big Data. IEEE, 2018.

This website provides information for the public on cycling in the Washington, DC area. Cyclists and others using this website assume all risk for injuries, damage, and other liabilities that may result from cycling. Bicycling can be a dangerous activity and regardless of the predicted safety information on this website, website users engage in cycling activities at their own risk. Users are advised to review the rules of the road, detailed maps of streets and trails, and other safety information prior to cycling. The University of Maryland, the State of Maryland, and their respective officers, agents, and employees are not responsible for any injury or other damage that results from use of this website.

Click here for the Cycling Safety project home page. For questions, please contact us at umdcylingsafety@gmail.com