Research on Imbalanced Data Classification and Computing Methods Using Support Vector Machine

In the current era of big data, all walks of life are transforming towards data-based operations. In this process, many scenarios often produce unevenly distributed data, which is often determined by the nature of the actual problem. Aiming at the problem of imbalanced data set classification, this paper proposes an improved Support Vector Machine model. After random sampling of most samples without replacement and sampling of minority samples with replacement to generate balanced training subsets, Support Vector Machine is trained to solve the problem of low accuracy of imbalanced data set prediction. Then we select the imbalanced data set on the public data set, and conduct an empirical analysis of our proposed scheme to verify the effectiveness of the improved Support Vector Machine model in solving the imbalanced data set classification problem from a relatively comprehensive perspective. The results show that the improved method proposed in this paper can significantly improve the classification performance of imbalanced data.