Analysis and Research on Spatial Layout of Urban Functional Elements Based on POI Data Analysis Based on Convolutional Neural Networks

For the study of the spatial layout of urban functional elements, a research method is proposed based on POI data and the convolutional neural network (CNN). The CNN uses the text information possessed by POI data and the text classification ability and combines the kernel density analysis method and natural breakpoint method in model connection. First, the POI data are pre-processed. Then, the convolutional neural network is used to classify the POI data into six categories: residential, public service, commercial, industrial, transportation, and cultural tourism. The classified data are processed by the kernel density analysis method and natural breakpoint method to obtain the spatial layout characteristics of urban functional elements and conduct relevant analysis. Taking Shenyang city as the research object, the following experimental results are obtained. (1) A research method of the spatial layout of urban functional elements is proposed based on POI data and a convolutional neural network to extract the spatial layout of urban functional elements more quickly than the traditional method. (2) Six functional elements are most densely distributed within the core urban area. (3) Commercial function and public service function, residential function. (4) There is a strong correlation between commercial function and public service function, residential function, and transportation function.