Method for deducing city functions based on communication data and interest point data

The invention relates to the field of urban planning. The invention aims to provide a method for deducing city functions based on communication data and interest point data. The method comprises the steps of dividing a research area into areas, superposing multi-source space-time big data with a division map, obtaining the number of internal features of each area segmented in the step 1, selectinga label sample of a training model, and establishing a deep learning model; deducing the urban function through the six steps of model training and urban function visualization according to the labelsamples obtained in the step 4. Future urban planning is facilitated, and the method has the advantages of being easy to implement, high in calculation speed, high in deduction precision and good invisualization effect. 本发明涉及城市规划领域，目的是提供一种基于通信数据与兴趣点数据推断城市功能的方法，本发明包括有将研究区划分区域、将多源时空大数据与划分地图叠加、获取步骤1中分割的每个区域内部特征的数量、选取训练模型的标签样本、建立深度学习模型，根据步骤4获得的标签样本对模型进行训练和城市功能可视化这6步对城市功能进行推断，便于对城市的未来规划，本发明具有实现简单，计算速度快，推断精度高，可视化效果好的优点。