摘要
基于多类别肿瘤基因表达谱数据集，从研究肿瘤与正常组织的分类入手，对肿瘤分类特征基因选取问题进行分析和研究。将决策树算法应用到肿瘤基因表达谱分类研究中，尝试引入遗传算法，对决策树分类规则进行优化。试验结果表明，在样本有限的情况下，该方法比单个决策树具有更高的分类精度。
The problem on analying the Multi-Class tumor gene expression profile datasets, and researching the classification of tumors and normal tissues start, tries to find the infor- mative genes for distinguishing tumor from normal tissues. Using the decision tree to the classification of tumor gene expression profiles, and tries to introduce genetic algorithm to optimize the classification rules derived by the decision tree. The result shows that this method can distinguish objects accurately and improve the precision.
作者
谢芬
XIE Fen （Computer Office of Teaching and Research, Binzhou Medical College, Binzhou 256600, China）
出处
《电脑知识与技术》
2010年第4期2493-2495,共3页
Computer Knowledge and Technology
基金
滨州医学院科技计划资助项目（BY2007KJ50）
关键词
肿瘤
遗传算法
肿瘤基因表达谱
决策树
特征基因
tumor
genetic algorithm
gene expression profiles
decision tree
informative gene
