中文摘要：

代价敏感普遍应用于解决分类不平衡问题,但代价敏感算法一直没有一个客观的评价标准.本文提出一种针对代价敏感算法的分类精度计算方法,以平衡精度替换总体精度来有效地评定代价敏感算法的分类性能.相比于传统的总体精度,该平衡精度不会忽略小类样本的贡献.通过代价敏感超限学习机对基因表达数据进行分类对比实验,结果表明,平衡精度可以更为客观、合理地表示代价敏感算法的分类性能.

英文摘要：

Cost sensitive algorithms are widely applied to solve the problem of unbalanced classification. However, there is no objective evaluation criteria for cost sensitive algorithms. This paper proposes a method of classification accuracy calculation for cost sensitive algorithms. Balance accuracy is utilized instead of overall accuracy to effectively assess the performance of cost sensitive algorithms. Compared with overall accuracy, the proposed balance accuracy will not neglect the contribution of samples in small classes. In the experiment, we classified gene expression data with cost sensitive extreme learning machines. The result shows the balance accuracy is a valid criterion for evaluating classification performance.