Extraction and monitoring approach of dynamic urban commercial area using check-in data from Weibo

Abstract

Urban hot spot detection and commercial area analysis for city economic planning and dynamic urban planning is of vital importance. However, it is difficult to obtain a more accurate commercial area boundary in the past. Check-in data obtained by a social networking service (SNS) and/or a location-based service (LBS) is a type of crowd-sourced geographic data that can reveal mass daily life activities, which provides a new big data source for urban hot spot detection and commercial area analysis. In this paper, a dynamic urban commercial area extraction and monitoring approach is proposed using SINA Weibo (a social network) check-in data. First, a check-in data pre-process model is proposed to simplify the amount of check-in data and improve the efficiency of cluster analysis. The spatial autocorrelation validation is implemented to validate the significant patterns of the spatial clustering of check-in data. Then, an exploratory spatial analysis and hot spot clustering method based on check-in data is proposed to detect urban hot spots and extract commercial areas using a geographic distribution metric with urban commercial hot spots. Second, the hot spot cluster analysis results are taken to determine the center of the commercial area and calculate the distribution of an ellipse, which is adopted to obtain the rough boundary of the commercial area. A planar Delaunay iterative triangulation algorithm is presented to determine the exact boundary of the commercial area. Then, the time sequence extraction result of the commercial area is presented to analyze the evolutionary trend in the city business space. Finally, the Weibo check-in data from 2012 to 2014 of Wuhan city are taken as an example dataset for the commercial area extraction and detection with the proposed approach. The results show that the method can accurately determine the boundary of and changes within the commercial area in Wuhan city. This study provides a new method for the monitoring of hot spots and the geographical situations of city commercial areas.