Interval-valued analysis for discriminative gene selection and tissue sample classification using microarray data

Genomics

Yunsong Qi¹  Xibei Yang¹ 

关键词: Microarray;    Gene selection;    Classification;    Rough sets;    Interval-valued decision table;   

DOI  :  10.1016/j.ygeno.2012.09.004

学科分类：医学（综合）

来源: Academic Press

【 摘 要 】

Animportantapplicationofgeneexpressiondataistoclassifysamplesinavarietyofdiagnosticfields.However,highdimensionalityandasmallnumberofnoisysamplesposesignificantchallengestoexistingclassificationmethods.Focusedontheproblemsofoverfittingandsensitivitytonoiseofthedatasetintheclassificationofmicroarraydata,weproposeaninterval-valuedanalysismethodbasedonaroughsettechniquetoselectdiscriminativegenesandtousethesegenestoclassifytissuesamplesofmicroarraydata.Wefirstselectasmallsubsetofgenesbasedoninterval-valuedroughsetbyconsideringthepreference-ordereddomainsofthegeneexpressiondata,andthenclassifytestsamplesintocertainclasseswithatermofsimilardegree.Experimentsshowthattheproposedmethodisabletoreachhighpredictionaccuracieswithasmallnumberofselectedgenesanditsperformanceisrobusttonoise.

【 授权许可】

Unknown   

【 预 览 】

附件列表

Files	Size	Format	View
RO201912090765622ZK.pdf	360KB	PDF