predictive analytics and machine learning

predictive analytics and machine learning

ID:7291051

大小:996.08 KB

页数:30页

时间:2018-02-10

predictive analytics and machine learning_第1页
predictive analytics and machine learning_第2页
predictive analytics and machine learning_第3页
predictive analytics and machine learning_第4页
predictive analytics and machine learning_第5页
资源描述:

《predictive analytics and machine learning》由会员上传分享,免费在线阅读,更多相关内容在工程资料-天天文库

1、PredictiveAnalyticsandMachineLearningPredictiveanalyticsandmachinelearningarehot,newresearchfields.Theyarenewcomparedtootherfieldsand,withoutadoubt,wecanexpectalotofrapidgrowth.Itisevenpredictedthatmachinelearningwillacceleratesofastthatwithinmeredecadeshumanlaborwillbereplacedbyintelligentmac

2、hines(seehttp://en.wikipedia.org/wiki/Technological_singularity).Thecurrentstateofartisfarfromthatutopia.Alotofcomputingpoweranddataisstillneededtomakeevensimpledecisions,suchasdeterminingwhetherpicturesontheInternetcontaindogsorcats.Predictiveanalyticsusesavarietyoftechniques,includingmachine

3、learningtomakeusefulpredictions,forinstance,todeterminewhetheracustomercanrepayhisorherloansoridentifyfemalecustomerswhoarepregnant(seehttp://www.forbes.com/sites/kashmirhill/2012/02/16/how-target-figured-out-a-teen-girl-was-pregnant-before-her-father-did/).Tomakethesepredictions,featuresareex

4、tractedfromhugevolumesofdata.Wementionedfeaturesbefore—theyarealsocalledpredictors.Featuresareinputvariablesthatcanbeusedtomakepredictions.Inessence,wehavefeaturesfoundinourdataandwearelookingforafunctionthatmapsthefeaturestoatarget,whichmayormaynotbeknown.Findingtheappropriatefunctioncanbehar

5、d;often,differentalgorithmsandmodelsaregroupedtogetherinsocalledensembles.Theoutputofanensemblecanbeamajorityvoteoranaverageofagroupofmodels,butwecanalsouseamoreadvancedalgorithmtoproducethefinalresult.Wewillnotbeusingensemblesinthischapter,butitissomethingtokeepinmind.PredictiveAnalyticsandMa

6、chineLearningInthepreviouschapter,wegotatasteofmachinelearningalgorithms—theNaiveBayesclassificationalgorithm.Wecandividemachinelearningintothefollowingcategories:•Supervisedlearning:Thisrequiresustolabeltrainingdata.Forinstance,ifwewanttoclassifyspam,weneedtoprovideexamplesofspamandnormale-ma

7、ilmessages.•Unsupervisedlearning:Thisdoesn'trequirehumaninput.Thistypeoflearningcandiscoverpatternssuchasclustersinlargedatasets.•Reinforcementlearning:Thisislearningwithoutatutor,butwithsomesortoffeedback.Forexample,acomputercanplayche

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。