Last modified on January 30, 2014, at 20:19

A utility for estimating the relative contributions of wiki authors

Publication (help)
A utility for estimating the relative contributions of wiki authors
Authors: Ofer Arazy, Eleni Stroulia [edit item]
Citation: Proceedings of the 3rd International Conference on Weblogs and Social Media (ICWSM’09)  : . 2009 May 2009. San Jose, California, USA.
Publication type: Conference paper
Peer-reviewed: Yes
DOI: Define doi.
Google Scholar cites: Citations
Link(s): Paper link
Added by Wikilit team: No
Article: Google Scholar BASE PubMed
Other scholarly wikis: AcaWiki Brede Wiki WikiPapers
Web search: Bing Google Yahoo!Google PDF
Format: BibTeX
A utility for estimating the relative contributions of wiki authors is a publication by Ofer Arazy, Eleni Stroulia.

[edit] Abstract

Wikis were originally designed to hide the association between a wiki page and the authors who have produced it. However, there is evidence suggesting that corporate wiki users require an attribution mechanism that would automatically record (and present) the relative contribution of each author. In this paper we introduce an algorithm for assessing the contributions of wiki authors that is based on the notion of sentence ownership. The results of an empirical evaluation comparing the algorithm’s output to manual evaluations reveal the type of contributions captured by our algorithm. Implications for research and practice are discussed.

[edit] Research questions

""there is a need for an attribution mechanism that would automatically record (and present) the relative contribution of each author (...) In this paper, we discuss our initial work towards addressing this concern, and introduce a wiki add-on that automatically calculates the relative contributions of wiki authors" (p. 171)"

Research details

Topics: Collaboration software [edit item]
Domains: Computer science [edit item]
Theory type: Design and action [edit item]
Wikipedia coverage: Case [edit item]
Theories: [edit item]
Research design: Statistical analysis [edit item]
Data source: Wikipedia pages [edit item]
Collected data time dimension: Longitudinal [edit item]
Unit of analysis: Edit, User [edit item]
Wikipedia data extraction: Live Wikipedia [edit item]
Wikipedia page type: History [edit item]
Wikipedia language: N/A [edit item]

[edit] Conclusion

""While non-attribution is useful in promoting democratic deliberation on the internet, it prevents corporate users from gaining recognition for their wiki work. Recent studies have proposed software utilities that would automatically attribute a wiki user with a score representing his contribution. However, these proposed algorithms suffer from several drawbacks, as they often use course measures, they are easy to manipulate, and they often capture just a sub-set of the classes of contributions. In this paper we’ve tried to address these gaps by proposing a novel wiki attribution algorithm and comparing it against human perceptions. The innovation of our algorithm lies in (a) the sentence-ownership algorithm, and (b) in calculating contributions that persist in the current version of the wiki page (in addition to metrics calculating the overall contribution). We argued that the count of sentences that survived the wiki process of continuous refinements implicitly captures the quality of a user’s contributions. One of the most surprising result of our study is that the metric that is most correlated with assessors’ perceptions of top contributors is the internal link count. We do not believe that assessors’ perceptions were strongly affected by the number of internal links an author makes. Rather, we explain this result by the fact that the ones adding links are active across a variety of categories, and this is why they are perceived as top contributors. We were also surprised to find that the simple edit count – used as a baseline – performed very well, yielding higher correlation with top contributor perceptions that other metrics such as sentence ownership. We believe that this is due to the fact that the edit count captures range of contributions across all categories, while the other metrics are associated with only a sub-set of the authoring categories. The sentence ownership metric performed fairly well, and has the advantage that it is less vulnerable to manipulations. Additional research is warranted in order to explore the design of more advanced wiki attribution algorithms, so that we can gain a better understanding of the authoring categories captured by various algorithms, and assess whether the presentation of user attribution indeed motivates wiki users to enhance their participation." (p. 174)"

[edit] Comments

Further notes[edit]