Research on C-end Customer Demand Based on Cluster Analysis

Abstract

With the rapid development of Internet economy, the problem of “last kilometer” distribution difficulty in B2C e-commerce has become one of the pain points of the whole industry, and improving the customer service level of C-end has become an important content of enterprise development. The change of C-end demand distribution will have a significant impact on the location of end-of-line service outlets, so this paper studies the spatial distribution of C-end customer demand points. According to the significant spatial clustering of C-end customer demand points in geographical space, K-means algorithm and DBSCAN algorithm are selected to cluster the demand points. The clustering results of the two algorithms are evaluated by the internal distance and the average distance between clusters. The comparison results show that k-means algorithm has better clustering effect on the demand points, and its clustering results can also be used as the end the main measurement index of service network layout provides services for terminal service network location.