摘要： 针对提取显现模式时在小样本情况下频率近似于概率的缺陷，在衡量分类信息能力熵的计算中引入贝叶斯方法估计概率P(Ci, Sj)，提高熵的可靠度，在此基础上提取癌症表达中的增强显现模式，提出2种基于增强显现模式的癌症分类算法。在急性白血病数据集上进行实验，结果表明，该算法能提高癌症检测的正确率。

关键词: 显现模式, 癌症分类, 基因表达模式

Abstract: For the defect of frequency similar to the probability when extracting Emerging Pattern(EP) in the case of small samples, Bayesian is introduced to evaluate the probability P(Ci, Sj) in measuring classified information capacity entropy for improving the reliability of entropy. It extracts Improved Emerging Pattern(IEP) from the cancer expression and gives two kinds of cancer classification algorithms based on IEP. Experiments are taken on the Acute Leukemia dataset and the results show the algorithm can improve the accuracy of cancer detection.

Key words: Emerging Pattern(EP), cancer classification, gene expression pattern

中图分类号: 

• TP181