基于改进K最近邻算法的中文文本分类.docxVIP

  • 15
  • 0
  • 约3.06千字
  • 约 6页
  • 2022-04-25 发布于北京
  • 举报

基于改进K最近邻算法的中文文本分类.docx

PAGE 1 - 基于改进K最近邻算法的中文文本分类 基于改良K最近邻算法的中文文本分类 5137〔2021〕01-0096-06 Abstract:Thispaperfocusesonthehighdimensionaltextproblemsencounteredintextclassification.Documentfrequency〔DF〕-chisquarestatisticfeatureextractionmethodisproposedtoreducethefeatureitemsandreducethedimensionoftext.BasedontheKNearestNeighbor〔KNN〕algorithm,inviewoftheproblemthattexttobeclassifiedshouldbecalculatedinsimilaritywithalargenumberoftrainingsetsamples,aKNNalgorithmbasedongroupingcentervectorisproposed.Thecentervectorsofeachgroupwereobtainedbygroupingthesamplesetsinthecategory,soastoimprovetheclassificationperformanc

文档评论(0)

1亿VIP精品文档

相关文档