Study on the classification of villages in Jilin Province based on space syntax and machine learning

Abstract

This paper employs machine learning and space syntax to classify villages, revealing implicit formation and development patterns in their spatial layouts. The study addresses village classification in Jilin Province from micro and macro perspectives. At the macro level, we present the classification method for villages based on geospatial distribution features (CVGD).We analyze the geographic distribution of villages and utilize village density for classification, effectively portraying the spatial distribution characteristics of villages. At the micro level, we propose the classification method for villages based on spatial structural features (CVSS). We use space syntax to extract spatial features of villages. Representative features chosen through comparative analysis serve as input variables in clustering algorithms, classifying villages based on high-dimensional data to explore their spatial traits. The paper summarizes village spatial characteristics from different classifications and studies how geographic factors affect village structure. It aims to offer potentially valuable theoretical insights for rural development. Abbreviations: POI: Point of Interest; DBSCAN: Density-Based Spatial Clustering of Applications with Noise; OPTICS: Ordering Points to Identify the Clustering Structure; WGS-84: World Geodetic System; ID: Identity Document; SSE: Sum of Squared Errors; CVGD: Classification Method for Villages Based on Geospatial Distribution Features; CVSS: Classification Method for Villages Based on Spatial Structure Features; GIS: Geographic Information Systems; ISUF: International Seminar on Urban Form; LiDAR: Light Detection and Ranging