A Study of Two-Phase Retrieval for两阶段检索研究

A Study of Two-Phase Retrieval for两阶段检索研究

ID:40843266

大小:631.23 KB

页数:22页

时间:2019-08-08

A Study of Two-Phase Retrieval for两阶段检索研究_第1页
A Study of Two-Phase Retrieval for两阶段检索研究_第2页
A Study of Two-Phase Retrieval for两阶段检索研究_第3页
A Study of Two-Phase Retrieval for两阶段检索研究_第4页
A Study of Two-Phase Retrieval for两阶段检索研究_第5页
资源描述:

《A Study of Two-Phase Retrieval for两阶段检索研究》由会员上传分享,免费在线阅读,更多相关内容在教育资源-天天文库

1、Chapter2AStudyofTwo-PhaseRetrievalforProcess-OrientedCase-BasedReasoningJosephKendall-MorwickandDavidLeakeAbstractProcess-OrientedCase-BasedReasoning(PO-CBR)systemsoftenusestructuredcases,whichinturnrequireeffectivestructure-basedretrievalmethods,especiallywhendealingwithl

2、argeprocessesand/orlargecasebases.Goodretrievalperformancecanbefacilitatedbytwo-phasedretrievalmethodswhichfirstwinnowcandidatecaseswithacomparativelyinexpensiveretrievalphase,andthenapplyamoreexpensivestrategytoranktheselectedcases.Examplesofsuchprocesseshavebeenshowntopro

3、videgoodretrievalresultsinlimitedretrievaltime.However,stud-iesofsuchmethodshavefocusedprimarilyonoverallperformance,ratherthanonhowtheindividualcontributionsofeachphaseinteracttoaffectoverallperformance.Thismissesanopportunitytotunethecomponentalgorithmsinlightofthesystem

4、taskandcasebasecharacteristics,forspecifictaskneeds.Thischapterexaminestwo-phasedretrievalasameansofaddressingthecomplexityinmanyPO-CBRdomains,andspecificallyexaminestheperformanceofeachphaseoftwo-phasedretrievalindividually,demonstratingcharacteristicsofthephases’interactio

5、nandprovidinggenerallessonsforhowtodesignanddeploytwo-phasedretrievalsystems.1IntroductionProcess-orientedtasks,suchasbusinessprocessmanagement,workflowgenera-tion,andplanning,requiregeneratingcomplexstructuredsolutions.Thegenerationofsuchsolutionsmaybedifficultduetoimperfec

6、tdomainknowledgeandcom-putationalcomplexity.Case-basedreasoning(CBR)(e.g.,[1–3])isappealingforJ.Kendall-Morwick(B)ComputerScienceDepartment,DePauwUniversity,Greencastle,IN46135,USAe-mail:josephkendallmorwick@depauw.eduD.LeakeSchoolofInformaticsandComputing,IndianaUniversit

7、y,Bloomington,IN47408,USAe-mail:leake@cs.indiana.eduS.MontaniandL.C.Jain(eds.),SuccessfulCase-basedReasoningApplications-2,7StudiesinComputationalIntelligence494,DOI:10.1007/978-3-642-38736-4_2,©Springer-VerlagBerlinHeidelberg20148J.Kendall-MorwickandD.Leakeaddressingbothp

8、roblems.BecauseCBRgeneratessolutionsbyrevisingthelessonsofrelevantpriorsolutions,itcaninc

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。