Modeling user reputation in wikis

From WikiLit
Jump to: navigation, search
Publication (help)
Modeling user reputation in wikis
Authors: Sara Javanmardi, Cristina Lopes, Pierre Baldi [edit item]
Citation: Statistical Analysis and Data Mining 3 (2): 126-139. 2010.
Publication type: Journal article
Peer-reviewed: Yes
Database(s):
DOI: 10.1002/sam.10070.
Google Scholar cites: Citations
Link(s): Paper link
Added by Wikilit team: Added on initial load
Search
Article: Google Scholar BASE PubMed
Other scholarly wikis: AcaWiki Brede Wiki WikiPapers
Web search: Bing Google Yahoo!Google PDF
Other:
Services
Format: BibTeX
Modeling user reputation in wikis is a publication by Sara Javanmardi, Cristina Lopes, Pierre Baldi.


[edit] Abstract

Collaborative systems available on the Web allow millions of users to share information through a growing collection of tools and platforms such as wikis, blogs, and shared forums. By their very nature, these systems contain resources and information with different quality levels. The open nature of these systems, however, makes it difficult for users to determine the quality of the available information and the reputation of its providers. Here, we first parse and mine the entire English Wikipedia history pages in order to extract detailed user edit patterns and statistics. We then use these patterns and statistics to derive three computational models of a user's reputation. Finally, we validate these models using ground-truth Wikipedia data associated with vandals and administrators. When used as a classifier, the best model produces an area under the receiver operating characteristic (ROC) curve (AUC) of 0.98. Furthermore, we assess the reputation predictions generated by the models on other users, and show that all three models can be used efficiently for predicting user behavior in Wikipedia.

[edit] Research questions

"Here, we first parse and mine the entire English Wikipedia history pages in order to extract detailed user edit patterns and statistics. We then use these patterns and statistics to derive three computational models of a user’s reputation. Finally, we validate these models using ground–truth Wikipedia data associated with vandals and administrators."

Research details

Topics: Vandalism, Reputation systems [edit item]
Domains: Computer science [edit item]
Theory type: Design and action [edit item]
Wikipedia coverage: Sample data [edit item]
Theories: "Undetermined" [edit item]
Research design: Statistical analysis [edit item]
Data source: Survey responses [edit item]
Collected data time dimension: Cross-sectional [edit item]
Unit of analysis: Edit, User [edit item]
Wikipedia data extraction: Live Wikipedia [edit item]
Wikipedia page type: Other [edit item]
Wikipedia language: English [edit item]

[edit] Conclusion

"In this paper, we have modeled user reputation in wiki systems. We have presented 3 reputation models. According to Model 1, when a user inserts a piece of content in a wiki page his reputation is increased and when a piece of content contributed by the user is deleted by another user, his reputation is decreased. In Model 2, we also take into account the time interval between insertions and deletions of content items. Finally, in Model 3, we add another parameter which is the current reputation of the deleter. Our experiments show that the three models can accurately assign reputation values to Wikipedia's known administrators/good users and vandals/blocked users. Additional analyses reveal that Model 1 does slightly better at detecting vandals and Model 3 does slightly better at detecting good users.

The proposed models can be applied in real time to calculate dynamic and individualized reputation values. While Model 1 is simpler to implement, Model 3 appears to be slightly more accurate, and more robust to attacks than several other competing models of reputation."

[edit] Comments


Further notes[edit]

Facts about "Modeling user reputation in wikis"RDF feed
AbstractCollaborative systems available on the WebCollaborative systems available on the Web allow millions of users to share information through a growing collection of tools and platforms such as wikis, blogs, and shared forums. By their very nature, these systems contain resources and information with different quality levels. The open nature of these systems, however, makes it difficult for users to determine the quality of the available information and the reputation of its providers. Here, we first parse and mine the entire English Wikipedia history pages in order to extract detailed user edit patterns and statistics. We then use these patterns and statistics to derive three computational models of a user's reputation. Finally, we validate these models using ground-truth Wikipedia data associated with vandals and administrators. When used as a classifier, the best model produces an area under the receiver operating characteristic (ROC) curve (AUC) of 0.98. Furthermore, we assess the reputation predictions generated by the models on other users, and show that all three models can be used efficiently for predicting user behavior in Wikipedia.for predicting user behavior in Wikipedia.
Added by wikilit teamAdded on initial load +
Collected data time dimensionCross-sectional +
ConclusionIn this paper, we have modeled user reputaIn this paper, we have modeled user reputation in wiki systems. We have presented 3 reputation models.

According to Model 1, when a user inserts a piece of content in a wiki page his reputation is increased and when a piece of content contributed by the user is deleted by another user, his reputation is decreased. In Model 2, we also take into account the time interval between insertions and deletions of content items. Finally, in Model 3, we add another parameter which is the current reputation of the deleter. Our experiments show that the three models can accurately assign reputation values to Wikipedia's known administrators/good users and vandals/blocked users. Additional analyses reveal that Model 1 does slightly better at detecting vandals and Model 3 does slightly better at detecting good users.

The proposed models can be applied in real time to calculate dynamic and individualized reputation values. While Model 1 is simpler to implement, Model 3 appears to be slightly more accurate, and more robust to attacks than several other competing models of reputation.
eral other competing models of reputation.
Data sourceSurvey responses +
Doi10.1002/sam.10070 +
Google scholar urlhttp://scholar.google.com/scholar?ie=UTF-8&q=%22Modeling%2Buser%2Breputation%2Bin%2Bwikis%22 +
Has authorSara Javanmardi +, Cristina Lopes + and Pierre Baldi +
Has domainComputer science +
Has topicVandalism + and Reputation systems +
Issue2 +
Pages126-139 +
Peer reviewedYes +
Publication typeJournal article +
Published inStatistical Analysis and Data Mining +
Research designStatistical analysis +
Research questionsHere, we first parse and mine the entire EHere, we first parse and mine the entire English

Wikipedia history pages in order to extract detailed user edit patterns and statistics. We then use these patterns and statistics to derive three computational models of a user’s reputation. Finally,

we validate these models using ground–truth Wikipedia data associated with vandals and administrators.
ssociated with vandals and administrators.
Revid10,876 +
TheoriesUndetermined
Theory typeDesign and action +
TitleModeling user reputation in wikis
Unit of analysisEdit + and User +
Urlhttp://dx.doi.org/10.1002/sam.10070 +
Volume3 +
Wikipedia coverageSample data +
Wikipedia data extractionLive Wikipedia +
Wikipedia languageEnglish +
Wikipedia page typeOther +
Year2010 +