Multi-perspective Visualization Analysis of Taxi Trajectory Data Features

The shortcomings of current analysis of taxi trajectory data include restricted perspectives and insufficient consideration of spatiotemporal features. To address these issues, this paper proposes a method for analyzing trajectory data from multiple angles to explore the information contained in taxi data in three dimensions: spatiotemporal hotspots, popular routes, and regional characteristics. Firstly, a new clustering algorithm is constructed, combining radial graphs and heat maps to analyze the travel characteristics of residents in hotspots. Then, spatial similarity is introduced to cluster hot routes, and urban travel characteristics are analyzed using hot route maps and flight maps. Finally, by calculating the ratio of weekend to weekday traffic flow, a radar chart of point of interest distribution is constructed to explore the composition characteristics of the city's three types of regions. The study found that roads such as Qiongzhou Avenue, Longhua Road, and Binhai Avenue in Haikou have high traffic volumes during rush hour, which easily leads to traffic congestion. Hotspot areas such as Xiuying Bridge, World Trade Center, and university areas have obvious changes, and the hotspot values are highest from 18:00 to 24:00 every day. The residential population activity areas are relatively stable, and people prefer to relax and seek medical treatment on weekends. The analysis shows that Haikou is a young city, and the experimental results are consistent with the age distribution of Haikou residents. The spatiotemporal characteristics of the city obtained from this research will provide valuable reference information for future urban planning and public facility layout.