The collaborative efforts of users in social media services such as Wikipedia have led to an explosion in user-generated content and how to automatically tag the quality of the content is an eminent concern now. Actually each article is usually undergoing a series of revision phases and the articles of different quality classes exhibit specific revision cycle patterns. We propose to Assess Quality based on Revision History (AQRH) for a specific domain as follows. First, we borrow Hidden Markov Model (HMM) to turn each article's revision history into a revision state sequence. Then, for each quality class its revision cycle patterns are extracted and are clustered into quality corpora. Finally, article's quality is thereby gauged by comparing the article's state sequence with the patterns of pre-classified documents in probabilistic sense. We conduct experiments on a set of Wikipedia articles and the results demonstrate that our method can accurately and objectively capture web article's quality. © 2011 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Han, J., Wang, C., & Jiang, D. (2011). Probabilistic quality assessment based on article’s revision history. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6861 LNCS, pp. 574–588). https://doi.org/10.1007/978-3-642-23091-2_50
Mendeley helps you to discover research relevant for your work.