Exploration of the Relationship between Travel Activity Chains and Self-Assessed Health of Rural Community Residents: Based on Multi-Source Data and Health Survey Questionnaire

Abstract

At present, most of China’s population still lives in rural areas, and the allocation of health facilities in rural areas is generally weaker than that in urban areas. The factors affecting the healthy life of rural residents need to be studied. This study took six rural communities in Pujiang County, Chengdu, Sichuan Province as the research object, and randomly selected 172 residents as samples to investigate the relationship between the characteristics of daily travel activities and self-rated health level of rural community residents, in order to provide a scientific basis for the construction of healthy rural communities and their surrounding supporting facilities. In view of the difficulty of data collection in rural areas, such studies are relatively rare in China. The research comprehensively uses multi-source data, including GPS data obtained by wearable positioning equipment, health questionnaire data, POI data of Amap and building vector data of MapWorld. Firstly, this study uses AT-DBSCAN algorithm to cluster the travel trajectory points to obtain the location information of residents’ stay points between 8:00-22:00, and divides the study area into 120m long grids. The TF-IDF algorithm is used to obtain the functional attributes of the grid based on the classified POI data, and then infer the type of stay point activities. For rural areas without POI data, the daily travel activity chain of rural community residents is constructed by combining building vector data. Then, the Word2Vec word embedding model is used to vectorize the activity chain data, and the K-means clustering analysis is used to obtain different daily travel activity patterns. Finally, combined with the residents’ self-rated health data obtained from the health questionnaire, the Nonparametric Tests was used to analyze the differences between the self-rated health levels of different travel activity modes. The results show that there are significant differences in self-rated health level among rural community residents with different travel activity modes, which provides a reference for rural community planning and health management.