Research on Boosting-based Imbalanced Data Classification
基于Boosting的不平衡数据分类算法研究

This paper aimed to investigate boosting-based unbalanced data classification algorithms. hhrough the deep analysis of existing algorithms, a weight sampling boosting algorithm was proposed. Changing the data distribution by weight sampling,the trained classifier was made suitable for unbalanced data classification. The natural of the proposed algorithm is that the loss function of naW c boosting is adjusted by the sampling function and the positive examples are emphasized so that the classifier focuses on correctly classifying these examples and finally the recognition rate of positive examples is improved. The new algorithm is simple and practical and has been shown to outperform naive boosting and previous algorithms in the problem of unbalanced data classification on the UCI data sets.