Integrating Frequent Pattern Mining from Multiple Data Domains for Classification

Integrating Frequent Pattern Mining from Multiple Data Domains for Classification

ID:40351668

大小:469.28 KB

页数:12页

时间:2019-07-31

Integrating Frequent Pattern Mining from Multiple Data Domains for Classification_第1页
Integrating Frequent Pattern Mining from Multiple Data Domains for Classification_第2页
Integrating Frequent Pattern Mining from Multiple Data Domains for Classification_第3页
Integrating Frequent Pattern Mining from Multiple Data Domains for Classification_第4页
Integrating Frequent Pattern Mining from Multiple Data Domains for Classification_第5页
资源描述:

《Integrating Frequent Pattern Mining from Multiple Data Domains for Classification》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、2012IEEE28thInternationalConferenceonDataEngineeringIntegratingFrequentPatternMiningfromMultipleDataDomainsforClassiſcationDhavalPatelWynneHsuMongLiLeeNationalUniversityofSingaporeSingapore{dhaval,whsu,leeml}@comp.nus.edu.sgAbstract—Manyfrequentpatternminingalg

2、orithmshavebeentomineusefulfrequentpatternsforclassiſcation[10],[24],developedforcategorical,numerical,timeseries,orinterval[14],[5],[6].data.However,littleattentionhasbeengiventointegratetheseAstraightforwardmethodtodiscoverheterogenouspatternsalgorithmssoasto

3、minefrequentpatternsinvolvingmultipleistoapplydifferentfrequentpatternminingalgorithmsforthedatadomainsforclassiſcation.Inthispaper,weintroducethenotionofaheterogenouspatternthatcapturestheassociationsdifferentdatadomains,followedbyanexhaustivecombina-amongdiff

4、erentdatadomains.Weproposeauniſedframeworktionsofthediscoveredpatterns.Aquickcalculationrevealsforminingmultipledomainsanddesignaniterativealgorithmthatthisapproachiscomputationallyinfeasible.AsmallcalledHTMiner.HTMinerdiscoversessentialheterogenouspat-datasetw

5、ith10categoricalattributes,20numericalattributes,ternsforclassiſcationandperformsinstanceelimination.This10events,and10daysof10timeseriesdatawouldresultinstanceeliminationstepreducestheproblemsizeprogressivelyinthegenerationof210frequentitemsets[22],220frequent

6、byremovingtraininginstanceswhicharecorrectlycoveredbythediscoveredessentialheterogenouspattern.Experimentsonintervals[9],1010frequenttemporalpatterns[14],and1010tworealworlddatasetsshowthattheHTMinerisefſcientandtimemotifspatterns[15].Thecombinationofthesepatte

7、rnsiscansigniſcantlyimprovetheclassiſcationaccuracy.58oftheorder2,andonlyasubsetofthesepatternsareusefulforclassiſcation.Clearly,weneedamoreintegratedapproachI.INTRODUCTIONtodiscoverusefulpatternsfromdifferentdatadomainsforeffectiveclassiſcation.Manydatabaseapp

8、licationsinvolverecordswithattributesEarlyworksonheterogenouspatternsarelimitedtominingfromdifferentdatadomains.Forexample,inaclinicalap-fromatmosttwodifferentkindsofdata[17

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。