A User Portrait of Express Software Based on Full Life Cycle Data
With the rapid development of traditional express industry, it is of great significance to study the portrait of express users. The express industry can realize the industrial transformation and upgrading by means of the express customer portrait. However, the research on the user portrait of express delivery is still scarce, at present. In this paper, a user portrait model based on life cycle data is proposed, which comprehensively describes the behavior preferences of users through multi-dimensional feature vectors. The model takes into account the basic attributes, preference attributes and feedback attributes of users, and uses the idea of classification before clustering to realize the segmentation of express users. In the classification stage, this paper fuses the K-means and the Support vector machine algorithm, adds the preference coefficient, and designs a new objective function to complete the user classification. In the process of clustering, we follow the idea of density-based peak clustering, combine the characteristic attributes of the multidimensional data of express users, add the characteristic coefficient, modify the index calculation formula in the model, and complete the clustering of express users. The simulation data show that the user portrait model proposed in this paper can achieve better segmentation of courier users, with better accuracy and timeliness.