会计数据分析 课件 第3章　模型建立与评估.pptx

ModelingandEvaluationChapter3

Wherewearenow1.DataAnalytics2.DataPreparationandCleaning3.ModelingandEvaluation4.Visualization5.TheModernAudit6.AuditAnalytics7.KeyPerformanceIndicators8.FinancialStatementAnalytics

ObjectivesLO3-1Whatdoyouneedtoknowaboutdatamodels?LO3-2Whataresomeunsupervisedandsupervisedapproaches?LO3-3Howdoyouperformprofiling?LO3-4Howdoyouperformdatareduction?LO3-5Howdoyouperformregression?LO3-6Howdoyouperformclassification?LO3-7Howdoyouperformclustering?

IntheIMPACTcycle,we’renowgoingtolookatPerformingtheTestPlanandAddressingtheResults.IdentifythequestionsMasterthedataPerformtestplanAddressandrefineresultsCommunicateinsightsTrackoutcomesExhibit1-1TheIMPACTCycle

Whatdoyouneedtoknowaboutdatamodels?LO3-1

Whatdoyouneedtoknowaboutdatamodels?Atargetisanexpectedattributeorvaluethatwewanttoevaluate.Example:FraudscoreInterestrateAclassisamanuallyassignedcategoryappliedtoarecordbasedonanevent.Example:Accept/RejectFraud/Notfraud

Whatdoyouneedtoknowfirst?Anunsupervisedapproachisusedwhenyoudon’thaveaspecificquestion.Example:“Doourvendorsformnaturalgroupsbasedonsimilarattributes?”Asupervisedapproachisusedwhenyouaretryingtopredictafutureoutcomebasedonhistoricaldata.Example:“Willanewvendorshipalargeorderontime?”

Whataresomeunsupervisedapproaches?Clustering–findundiscoverednaturalgroupingsinthedataCo-occurrencegrouping–eventsthathappentogetherProfiling–identifytypicalbehaviorinthedataDatareduction–filterorgroupthedatatosimplifytheanalysis

Whataresomesupervisedapproaches?Classification–predictwhetherdatabelongstooneclassoranotherSimilaritymatching–groupdatabyattributesRegression–predictaspecificvalueLinkprediction–socialnetworksCausalmodeling–aneventinfluencesanother

Useaflowcharttoident