Modeling and data mining in blogosphere

113 Pages·2009·3.216 MB·English

by  Nitin Agarwal, Huan Liu

Modeling and Data Mining
in Blogosphere

Synthesis Lectures on Data Mining
and Knowledge Discovery
Editor
RobertGrossman,UniversityofIllinois,Chicago
ModelingandDataMininginBlogosphere
NitinAgarwalandHuanLiu
2009
Copyright© 2009byMorgan&Claypool
Allrightsreserved.Nopartofthispublicationmaybereproduced,storedinaretrievalsystem,ortransmittedin
anyformorbyanymeans—electronic,mechanical,photocopy,recording,oranyotherexceptforbriefquotationsin
printedreviews,withoutthepriorpermissionofthepublisher.
ModelingandDataMininginBlogosphere
NitinAgarwalandHuanLiu
www.morganclaypool.com
ISBN:9781598299083 paperback
ISBN:9781598299090 ebook
DOI10.2200/S00213ED1V01Y200907DMK001
APublicationintheMorgan&ClaypoolPublishersseries
SYNTHESISLECTURESONDATAMININGANDKNOWLEDGEDISCOVERY
Lecture#1
SeriesEditor:RobertGrossman,UniversityofIllinois,Chicago
SeriesISSN
SynthesisLecturesonDataMiningandKnowledgeDiscovery
ISSNpending.
Modeling and Data Mining
in Blogosphere
Nitin Agarwal
UniversityofArkansasatLittleRock
Huan Liu
ArizonaStateUniversity
SYNTHESISLECTURESONDATAMININGANDKNOWLEDGEDISCOVERY
#1
M
&C Morgan &cLaypool publishers
ABSTRACT
Thisbookoffersacomprehensiveoverviewofthevariousconceptsandresearchissuesaboutblogsor
weblogs.It introduces techniques and approaches,tools and applications,and evaluation method-
ologies with examples and case studies. Blogs allow people to express their thoughts, voice their
opinions,andsharetheirexperiencesandideas.Blogsalsofacilitateinteractionsamongindividuals
creating a network with unique characteristics.Through the interactions individuals experience a
senseofcommunity.Weelaborateonapproachesthatextractcommunitiesandclusterblogsbased
on information of the bloggers. Open standards and low barrier to publication in Blogosphere
have transformed information consumers to producers, generating an overwhelming amount of
ever-increasingknowledgeaboutthemembers,theirenvironmentandsymbiosis.Weelaborateon
approaches that sift through humongous blog data sources to identify influential and trustworthy
bloggersleveragingcontentandnetworkinformation.Spamblogsorsplogsisanincreasingconcern
in Blogosphere, which is discussed in detail with the approaches leveraging supervised machine
learning algorithms and interaction patterns.We elaborate on data collection procedures,provide
resourcesforblogdatarepositories,mentionvariousvisualizationandanalysistoolsinBlogosphere,
andexplainconventionalandnovelevaluationmethodologies,tohelpperformresearchintheBlo-
gosphere.
Thebookissupportedbyadditionalmaterial,includinglectureslidesaswellasthecomplete
setoffiguresusedinthebook,andthereaderisencouragedtovisitthebookwebsiteforthelatest
information:
http://tinyurl.com/mcp-agarwal
KEYWORDS
blogosphere,weblogs,blogs,blog model,power law distribution,scale free networks,
degree distribution,clustering coefficient,centrality measures,clustering,community
discovery,influence,diffusion,trust,propagation,spam blogs,splogs,data collection,
blogcrawling,performanceevaluation
To my parents, Sushma and Umesh Chand Agarwal…–NA
To my parents, wife, and sons…–HL
…with much love and gratitude for everything.

ix
Contents
Acknowledgments................................................................xi
1 ModelingBlogosphere............................................................1
1.1 ModelingEssentials ........................................................2
1.2 PreferentialAttachmentBlogModels ........................................ 8
1.2.1 Log-normalDistributionModels 12
2 BlogClusteringandCommunityDiscovery........................................15
2.1 GraphBasedApproach....................................................17
2.2 ContentBasedApproach...................................................21
2.3 HybridApproach..........................................................24
3 InfluenceandTrust..............................................................27
3.1 Influence .................................................................27
3.1.1 GraphBasedApproach 30
3.1.2 ContentBasedApproach 33
3.1.3 HybridApproach 34
3.1.4 BlogLeaders 40
3.2 Trust.....................................................................40
3.2.1 TrustComputation 41
3.2.2 TrustPropagation 43
4 SpamFilteringinBlogosphere....................................................45
4.1 GraphBasedApproach....................................................47
4.2 ContentBasedApproach...................................................49
4.3 HybridApproach..........................................................51
5 DataCollectionandEvaluation...................................................53
x CONTENTS
5.1 DataCollection...........................................................53
5.1.1 API 53
5.1.2 WebCrawler 56
5.1.3 AvailableDatasets 58
5.1.4 DataPreprocessing 59
5.2 Evaluation................................................................60
5.2.1 BlogModeling 61
5.2.2 BlogClusteringandCommunityDiscovery 61
5.2.3 InfluenceandTrust 64
5.2.4 Spam 68
A ToolsinBlogosphere.............................................................71
B APIExamples...................................................................79
Bibliography ....................................................................87
Biography.......................................................................95
Index...........................................................................97