Firstly, the historical data of domestic road traffic incidents are extracted, and a classification model based on K-means clustering algorithm is constructed from four aspects: death toll, number of disabled people, economic loss and impact. Divide the risk of traffic incidents from high to low into four levels: A-level specially serious risk, B-level large risk, C-level general risk and D-level minor risk. Then, using big data analysis techniques, Using the angle cosine model to compare the similarity of the traffic incident just happened attribute vectors with the attribute values of the different risk levels in the database to determine the A, B, C, and D levels of the traffic incident just happened. Finally, using the constructed geographic information system dedicated to public security traffic, analysis the temporal and spatial characteristics of police resources and the traffic incident, and automatically distribute the low-risk traffic incident alert to the road patrol policemen who is closest to the traffic incident in the jurisdiction to achieve rapid response, the response efficiency can be doubled.

CICTP