- 1、本文档共10页,可阅读全部内容。
- 2、原创力文档(book118)网站文档一经付费(服务费),不意味着购买了该文档的版权,仅供个人/单位学习、研究之用,不得用于商业用途,未经授权,严禁复制、发行、汇编、翻译或者网络传播等,侵权必究。
- 3、本站所有内容均由合作方或网友上传,本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺!文档内容仅供研究参考,付费前请自行鉴别。如您付费,意味着您自己接受本站规则且自行承担风险,本站不退款、不进行额外附加服务;查看《如何避免下载的几个坑》。如果您已付费下载过本站文档,您可以点击 这里二次下载。
- 4、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等,请点击“版权申诉”(推荐),也可以打举报电话:400-050-0827(电话支持时间:9:00-18:30)。
RetrievalEvaluation医疗专题知识
Outline
Introduction
RetrievalPerformanceEvaluation
Recallandprecision
Alternativemeasures
ReferenceCollections
TRECCollection
CACMISICollection
CFCollection
TrendsandResearchIssues
Introduction
Typeofevaluation
Functionalanalysisphase,andErroranalysisphase
Performanceevaluation
Performanceevaluation
Responsetime/spacerequired
Retrievalperformanceevaluation
Theevaluationofhowpreciseistheanswerset
RetrievalPerformanceEvaluation
評估以batchquery為主旳IR系統
collection
RelevantDocs
InAnswerSet
|Ra|
RelevantDocs
|R|
AnswerSet
|A|
Recall=|Ra|/|R|
Precision=|Ra|/|A|
Sortedbyrelevance
Precisionversusrecallcurve
Rq={d3,d5,d9,d25,d39,d44,d56,d71,d89,d123}
P=100%atR=10%
P=66%atR=20%
P=50%atR=30%
Rankingforqueryq:
Usuallybasedon11standardrecalllevels:0%,10%,...,100%
Precisionversusrecallcurve
Forasinglequery
Fig3.2
AverageOverMultipleQueries
P(r)=averageprecisionattherecalllevelr
Nq=Numberofqueriesused
Pi(r)=Theprecisionatrecalllevelrforthei-thquerInterpolatedprecision
Rq={d3,d56,d129}
P=33%atR=33%
P=25%atR=66%
P=20%atR=100%
P(rj)=maxri≦r≦rj+1P(r)
Interpolatedprecision
Letrj,j{0,1,2,…,10},beareferencetothej-thstandardrecalllevel
P(rj)=maxri≦r≦rj+1P(r)
R=30%,P3(r)~P4(r)=33%
R=40%,P4(r)~P5(r)
R=50%,P5(r)~P6(r)
R=60%,P6(r)~P7(r)=25%
Averagerecallvs.precisionfigure
SingleValueSummaries
Averageprecisionversusrecall:
Compareretrievalalgorithmsoverasetofexamplequeries
Sometimesweneedtocompareindividualquery’sperformance
Averageprecision可能會隱藏演算法中不正常旳部分
可能需要懂得,兩個演算法中,對某特定query旳performance為何
Needasinglevaluesummary
Thesinglevalueshouldbeinterpretedasasummaryofthecorrespondingprecisionversusrecallcurve
SingleValueSummaries
AveragePrecisionatSeenRelevantDocuments
Averagingtheprecisionfiguresobtainedaftereachnewrelevantdocumentisobserved.
Example:Figure3.2,(1+0.66+0
文档评论(0)