Community Mining from Signed Social Networks

Community Mining from Signed Social Networks

ID:39785784

大小:4.99 MB

页数:16页

时间:2019-07-11

Community Mining from Signed Social Networks_第1页
Community Mining from Signed Social Networks_第2页
Community Mining from Signed Social Networks_第3页
Community Mining from Signed Social Networks_第4页
Community Mining from Signed Social Networks_第5页
资源描述:

《Community Mining from Signed Social Networks》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、IEEETRANSACTIONSONKNOWLEDGEANDDATAENGINEERING,VOL.19,NO.10,OCTOBER20071333CommunityMiningfromSignedSocialNetworksBoYang,WilliamK.Cheung,andJimingLiuAbstract—Manycomplexsystemsintherealworldcanbemodeledassignedsocialnetworksthatcontainbothpositiveandnegativerelations.Algorithmsforminingsocia

2、lnetworkshavebeendevelopedinthepast;however,mostofthemweredesignedprimarilyfornetworkscontainingonlypositiverelationsand,thus,arenotsuitableforsignednetworks.Inthiswork,weproposeanewalgorithm,calledFEC,tominesignedsocialnetworkswherebothpositivewithin-grouprelationsandnegativebetween-groupr

3、elationsaredense.FECconsidersboththesignandthedensityofrelationsastheclusteringattributes,makingiteffectivefornotonlysignednetworksbutalsoconventionalsocialnetworksincludingonlypositiverelations.Also,FECadoptsanagent-basedheuristicthatmakesthealgorithmefficient(inlineartimewithrespecttothes

4、izeofanetwork)andcapableofgivingnearlyoptimalsolutions.FECdependsononlyoneparameterwhosevaluecaneasilybesetandrequiresnopriorknowledgeonhiddencommunitystructures.TheeffectivenessandefficacyofFEChavebeendemonstratedthroughasetofrigorousexperimentsinvolvingbothbenchmarkandrandomlygeneratedsig

5、nednetworks.IndexTerms—Communitymining,agent-basedapproach,randomwalk,signedsocialnetworks.Ç1INTRODUCTIONANYcomplexsystemsintherealworldcanbemodeledthedensityoflinksbutalsobythesignsoflinks.Massocialnetworks[1],[2].AsocialnetworkcanbeQualitativelyspeaking,theyaredefinedasthegroupsofdefineda

6、sagraphG¼ðV;EÞ,whereV¼fv1;...;vngisthenodeswithinwhichthepositivelinksaredenseandsetofvertices,andEisthesetofedgesconnectingpairsofbetweenwhichnegativelinksarealsodense.Thechallengeofminingsignednetworkcommunitiesliesinthefactthatitvertices.Forexample,inahumansocialnetwork,eachisnaturaltoha

7、vesomenegativelinkswithingroupsand,atvertexdenotesanindividualoranode,andeachedgethesametime,somepositivelinksbetweengroups.Findingdenotesarelationoralinkbetweentwonodes.Inweightedthebestandyetrobustpartitionofthenetworkforsocialnetworks,eachlinkisattach

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。