全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

基于g方检验的cp-nets学习

DOI: 10.13232/j.cnki.jnju.2015.04.016, PP. 781-795

Keywords: g方检验,对数似然比检验,因果关系,条件偏好无关,零假设检验

Full-Text   Cite this paper   Add to My Lib

Abstract:

偏好处理是人工智能中的一个重要研究内容。cp-nets(conditionalpreferencenetworks,条件偏好网)是一个带标记的有向图,它编码相关变量之间的偏好关系。作为一种简单直观的图形偏好表示工具,却很少有工作对cp-nets的结构进行研究。研究cp-nets的结构,提出了基于g方检验对cp-nets进行结构学习的算法,并给出算法的时间复杂度为o(n?2n).作为一种对数似然比检验方法,g方检验特别适合于判断变量之间的因果关系。由于cp-nets的核心概念是条件偏好无关,因此利用g方检验可有效地实现cp-nets的结构学习。通过构造g方检验的统计量,在给定的成对比较样本集中,执行零假设检验,从而依次求出每个顶点的父亲集,进而得到cp-nets的结构。最后,通过随机生成的模拟数据,验证了所提出算法的有效性。与相关cp-nets的学习算法对比,本文提出的方法具有被动的,离线的,和基于统计学习的特征。

References

[1]  .colombod,maathuismh.order-independentconstraint-basedcausalstructurelearning.journalofmachinelearningresearch,2014,15(1):3741-3782.
[2]  .boutilierc,brafmanr,domshlakc,etal.cp-nets:atoolforrepresentingandreasoningwithconditionalceterisparibuspreferencestatements.journalofartificialintelligenceresearch,2004,21(1):135-191.
[3]  .lut,boutilierc.effectivesamplingandlearningformallowsmodelswithpairwisepreferencesdata.journalofmachinelearningresearch,2014,16(12):3783-3829.
[4]  .pomyeny,seguram,ebbelstm,etal.over-representationofcorrelationanalysis(orca):amethodforidentifyingassociationsbetweenvariablesets.bioinformatics,2015,31(1):102-108.
[5]  .liuw,wuc,fengb,etal.conditionalpreferenceinrecommendersystems.expertsystemswithapplications,2015,42(2):774-788.
[6]  .dimopoulosy,michaell,athienitouf.ceterisparibuspreferenceelicitationwithpredictiveguarantees.in:proceedingsofthe21thinternationaljontconferenceonartificialintelligence.sanfrancisco,ca,usa:morgankaufmannpublishersinc.,2009:1890-1895.
[7]  .langj,menginj.thecomplexityoflearningseparableceterisparibuspreferences.in:proceedingsofthe21thinternationaljontconferenceonartificialintelligence.sanfrancisco,ca,usa:morgankaufmannpublishersinc.,2009:848-853.
[8]  .langj,menginj.learningpreferencerelationsovercombinatorialdomains.in:hüllermeiere,fürnkranzj.in:proceedingsoftheworkshoponpreferencelearningattheeuropeanconferenceonmachinelearning(pl’08).antwerpen,belgium,2008.
[9]  .liuj,yaoz,xiongy,etal.learningconditionalpreferencenetworkfromnoisysamplesusinghypothesistesting.knowledge-basedsystems,2013,40(0):7-16.
[10]  .cornelioc,goldsmithj,mattein,etal.updatesanduncertaintyincp-nets.in:cranefields,nayaka.proceedingsofthe26thaustralasianjointconferenceonartificialintelligence.dunedin,newzealand:springerinternationalpublishing,2013:301-312.
[11]  .brafmanr,domshlakc.preferencehandling.aimagazine,2009,30(1):58-86.
[12]  .ailonn.learningandoptimizingwithpreferences.in:jains,munosr,stephanf,etal.algorithmiclearningtheory.germany:springerberlinheidelberg,2013:13-21.
[13]  .shiy,larsonm,hanjalica.collaborativefilteringbeyondtheuser-itemmatrix:asurveyofthestateoftheartandfuturechallenges.acmcomputersurvey,2014,47(1):3:1-3:45.
[14]  .韦素云,业宁,吉根林等.基于项目类别和兴趣度的协同过滤推荐算法.南京大学学报(自然科学),2013,49(2):142-149.
[15]  .conitzerv.makingdecisionsbasedonthepreferencesofmultipleagents.communicationsoftheacm,2010,53(3):84-94.
[16]  .刘惊雷.cp-nets及其表达能力研究.自动化学报,2011,37(3):290-302.
[17]  .王红兵,孙文龙,王华兰.web服务选择中偏好不确定问题的研究.计算机学报,2013,36(2):275-285.
[18]  .jensenfv,nielsentd.bayesiannetworksanddecisiongraphs(secondedition).berlin,germany:springerverlag,2007.
[19]  .hüllermeiere,fürnkranzj.editorial:preferencelearningandranking.machinelearning,2013,93(2-3):185-189.
[20]  .ailonn,charikarm,newmana.aggregatinginconsistentinformation:rankingandclustering.acm,2008,55(5):123-128.
[21]  .korichef,zanuttinib.learningconditionalpreferencenetworks.artificialintelligence,2010,174(11):685-703.
[22]  .ailonn.anactivelearningalgorithmforrankingfrompairwisepreferenceswithanalmostoptimalquerycomplexity.journalofmachinelearningresearch,2012,13(1):137-164.
[23]  .mcdonaldj.handbookofbiologicalstatistics.sparkyhousepublishing,2008.
[24]  .dalyr,shenq,aitkens.learningbayesiannetworks:approachesandissues.theknowledgeengineeringreview,2011,26(2):99-157.
[25]  .chickeringdm.learningequivalenceclassesofbayesian-networkstructures.journalofmachinelearningresearch,2002,2(3):445-498.
[26]  .waegemanw,dembczynskik,jachnika,etal.onthebayes-optimalityoff-measuremaximizers.journalofmachinelearningresearch,2014,15(10):3333-3388.

Full-Text

Contact Us

[email protected]

QQ:3279437679

WhatsApp +8615387084133