浅探关节镜下盘状半月板损伤治疗PPT课件.pptx

CS276B

TextInformationRetrieval,Mining,andExploitationLecture523January20031

Recap2

Today’stopicsFeatureselectionfortextclassificationMeasuringclassificationperformanceNearestneighborcategorization3

FeatureSelection:Why?Textcollectionshavealargenumberoffeatures10,000–1,000,000uniquewords–andmoreMakeusingaparticularclassifierfeasibleSomeclassifierscan’tdealwith100,000soffeat’sReducetrainingtimeTrainingtimeforsomemethodsisquadraticorworseinthenumberoffeatures(e.g.,logisticregression)ImprovegeneralizationEliminatenoisefeaturesAvoid