基于层级规则树的跨平台基因表达数据分类  

Cross-platform Gene Expression Data Classification Based on Hierarchical Rule Tree

作　　者：蔡瑞初[1] 侯永杰 郝志峰[1,2] CAI Ruichu;HOU Yongjie;HAO Zhifeng(School of Computers,Guangdong University of Technology,Guangzhou 510006,China;School of Mathematics and Big Data,Foshan University,Foshan,Guangdong 528000,China)

机构地区：[1]广东工业大学计算机学院,广州510006 [2]佛山科学技术学院数学与大数据学院,广东佛山528000

出　　处：《计算机工程》2019年第7期26-31,共6页Computer Engineering

基　　金：国家自然科学基金(61472089);NSFC-广东联合基金(U1501254);广东省自然科学基金(2014A030306004,2014A030308008);广东省科技计划项目(2015B010108006,2015B010131015);广东特支计划(2015TQ01X140);广州市科技计划项目(201604016075);广州市珠江科技新星专项(201610010101)

摘　　要：基因检测技术运用至今已积累大量来自不同平台的数据,针对传统数据分类模式难以在不同平台间进行有效迁移的问题,提出一种基于层级规则树的基因表达数据分类算法k-HRT。设计数据转换与规则预筛选策略,实现算法的快速挖掘,以解决由跨平台特性所带来的大规模数据问题。在真实基因表达数据集上的实验结果表明,相对k-TSP算法、SVM-RFE算法,k-HRT算法能够有效提高分类精度。The application of genetic testing technology has accumulated a large amount of data from different platforms.To address the problem that it is difficult to migrate traditional data classification modes across different platforms,this paper proposes a gene expression data classification algorithm k-HRT based on Hierarchy Rule Tree(HRT).The strategy of data conversion and rule pre-screening is designed to realize the fast mining of the algorithm to solve the large-scale data problems caused by cross-platform characteristics.Experimental results on real gene expression datasets show that,compared with k-TSP algorithm and SVM-RFE algorithm,k-HRT algorithm can effectively improve classification accuracy.

关 键 词：数据分类 跨平台 规则学习 特征选择 基因表达数据 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]