Hot Spot Area Identification of Online Car-Hailing Based on RNN-DBSCAN

The identification of hot spot areas of online car hailing can provide support for improving the operational efficiency of online car hailing. Taking the order data of Didi Express in Chengdu as samples, the reverse nearest neighbor clustering algorithm (RNN-DBSCAN) process, including travel order data processing, location coordinate transformation, and travel hotspot identification, is designed. Based on the sample data, the empirical analysis of the locations of hot spots and centers of online ride-hailing trips at different times of working days and weekends within the research scope is given. The results show that the improved algorithm overcomes the shortcomings of the traditional DBSCAN algorithm, which has arbitrary parameter Settings and cannot distinguish large density variable clusters, and provides a feasible scheme for data-driven ride-hailing travel characteristics analysis.