Wikipedia is a very large and successful Web 2.0 example. As the number of Wikipedia articles and contributors grows at a very fast pace, there are also increasing disputes occurring among the contributors. Disputes often happen in articles with controversial content. They also occur frequently among contributors who are “aggressive” or controversial in their personalities. In this paper, we aim to identify controversial articles in Wikipedia. We propose three models, namely the Basic model and two Controversy Rank (CR) models. These models draw clues from collaboration and edit history instead of interpreting the actual articles or edited content. While the Basic model only considers the amount of disputes within an article, the two Controversy Rank models extend the former by considering the relationships between articles and contributors. We also derived enhanced versions of these models by considering the age of articles. Our experiments on a collection of 19,456 Wikipedia articles shows that the Controversy Rank models can more e?ectively determine controversial articles compared to the Basic and other baseline models.

Based on their previous work, the authors of this paper initially propose three Controversy Ranking (CR) models to identify controversial Wikipedia articles. For example, 'Holiest sites in Islam' is considered as a very controversial Wikipedia article and has attracted disputes between authors in agreeing on the list of sites and in which rank they should be ordered. Currently, Wikipedia allows users to manually assign controversy related tags to articles (i.e. {{disputed}}, {{controversial}}, {{pov}} etc...). The problems with this manual approach is that all articles may not be reviewed by users for controversy and there may be controversial articles that are untagged. Moderators and contributers can improve the effectiveness of dispute resolution processes by being able to quickly identify controversial articles.

The results generated from the authors' CR models was tested against a data set of 19,456 articles from the Religious Objects category (considered to be a highly controversial topic category) in Wikipedia. Two evaluation metrics were adopted to evaluate these models which included the Precision-Recall-F1 at top k and the Normalized Discounted Cumulative Gain at top k (NDCG@k). However, the NDCG@k took into account different degrees of controversy in articles and was therefore considered a more accurate measure. Based on this measure, it was identified that the CR Product and CR Average models yielded the best results in identifying and ranking controversial articles.

However, the authors realised that the CR models did not take into account the age of the articles. Newly created articles can undergo many revisions amongst various contributors in disputing the organisation and presentation of article content whilst attempting to form a consensus. Articles that undergo numerous revisions (disputes) by contributors may be mistakenly identified as controversial even though the article itself is not controversial. Therefore the CR models were refined to be age-aware so these new and non-controversial articles would not be flagged as controversial. The results showed that the age-aware CR models significantly outperformed the non-age-aware CR models and yielded excellent results. Personally, from this paper, I am interested in learning more about the NDCG@k performance evaluation metric and how I can apply in my research.

• Controversy rank models-aware search
• Article & contributor controversy
• Evolving content
• Article disputed tags
• Contribution and dispute extraction
• Information retrieval (IR) performance measures
• Age-aware models
• Support vector machine (SVM) classifier