Multi-label Learning with User Credit Data in China Based on MLKNN

Abstract

Aiming at the problem of numerous variables in credit data, a large amount of sample data, and inability to intuitively reflect user portraits. This paper uses the MLKNN algorithm to perform multi-label learning on the credit data. According to the results of the algorithm training under 24 sets of k values, the optimal number of neighbor samples of the algorithm on the sample set is 14. On this basis, this paper further analyzes The general portrait of credit users in my country has the following characteristics: more than 50% of credit users are users with stable personal development, low frequency of credit activities, and low-to-medium attention to credit status. Meanwhile, we find that the users with higher frequency of credit activities pay more attention to credit status. This research can provide some reference for commercial banks or other financial institutions in lending and credit management.