资源描述:
《Machine Learning in Action.pdf》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库。
1、INACTIONPeterHarringtonMANNINGMachineLearninginActionMachineLearninginActionPETERHARRINGTONMANNINGShelterIslandForonlineinformationandorderingofthisandotherManningbooks,pleasevisitwww.manning.com.Thepublisheroffersdiscountsonthisbookwhenorderedinquantity.Formoreinfor
2、mation,pleasecontactSpecialSalesDepartmentManningPublicationsCo.20BaldwinRoadPOBox261ShelterIsland,NY11964Email:orders@manning.com©2012byManningPublicationsCo.Allrightsreserved.Nopartofthispublicationmaybereproduced,storedinaretrievalsystem,ortransmitted,inanyformorb
3、ymeanselectronic,mechanical,photocopying,orotherwise,withoutpriorwrittenpermissionofthepublisher.Manyofthedesignationsusedbymanufacturersandsellerstodistinguishtheirproductsareclaimedastrademarks.Wherethosedesignationsappearinthebook,andManningPublicationswasawareofa
4、trademarkclaim,thedesignationshavebeenprintedininitialcapsorallcaps.Recognizingtheimportanceofpreservingwhathasbeenwritten,itisManning’spolicytohavethebookswepublishprintedonacid-freepaper,andweexertourbesteffortstothatend.Recognizingalsoourresponsibilitytoconserveth
5、eresourcesofourplanet,Manningbooksareprintedonpaperthatisatleast15percentrecycledandprocessedwithouttheuseofelementalchlorine.ManningPublicationsCo.Developmenteditor:JeffBleiel20BaldwinRoadTechnicalproofreaders:TriciaHoffman,AlexOttPOBox261Copyeditor:LindaRecktenwald
6、ShelterIsland,NY11964Proofreader:MaureenSpencerTypesetter:GordanSalinovicCoverdesigner:MarijaTudorISBN9781617290183PrintedintheUnitedStatesofAmerica12345678910–MAL–171615141312ToJosephandMilobriefcontentsPART1CLASSIFICATION............................................
7、...................11■Machinelearningbasics32■Classifyingwithk-NearestNeighbors183■Splittingdatasetsonefeatureatatime:decisiontrees374■Classifyingwithprobabilitytheory:naïveBayes615■Logisticregression836■Supportvectormachines1017■ImprovingclassificationwiththeAdaBoos
8、tmeta-algorithm129PART2FORECASTINGNUMERICVALUESWITHREGRESSION..............1518■Predictingnumericvalues:regression1539■Tree-basedre