A Study on Functional Planning in Shanghai Regions Using K-Means and DBSCAN Clustering

Urban development tends to be complex and diverse, so planning theories can no longer meet the needs of urban development. Big data provides a new direction for the development of urban planning, using social network data. Through the use of the K-means clustering algorithm we quickly divide the city area, and through the DBSCAN clustering algorithm we get any shape of clusters in the divided area and can get their centers by calculating the average coordinates. Finally, the housing prices of each district are taken as the main line and combined with GDP data to analyze each district. The result shows that under the overall prosperity of Shanghai, there is a significant development gap in some regions. The results of this paper can provide some references for Shanghai's urban planning recommendations.