Algorithms and criteria for diversification of news article comments

In this paper, we introduce an approach for diversifying user comments on news articles. We claim that, although content diversity suffices for the keyword search setting, as proven by existing work on search result diversification, it is not enough when it comes to diversifying comments of news articles. Thus, in our proposed framework, we define comment-specific diversification criteria in order to extract the respective diversification dimensions in the form of feature vectors. These criteria involve content similarity, sentiment expressed within comments, named entities, quality of comments and combinations of them. Then, we apply diversification on comments, utilizing the extracted features vectors. The outcome of this process is a subset of the initial set that contains heterogeneous comments, representing different aspects of the news article, different sentiments expressed, different writing quality, etc. We perform an experimental analysis showing that the diversity criteria we introduce result in distinctively diverse subsets of comments, as opposed to the baseline of diversifying comments only w.r.t. to their content. We also present a prototype system that implements our diversification framework on news articles comments