Detecting Comments on News Articles in Microblogs

Abstract

A reader of a news article would often be interested in the comments of other readers on an article, because comments give insight into popular opinions or feelings toward a given piece of news. In recent years, social media platforms, such as Twitter, have become a social hub for users to communicate and express their thoughts. This includes sharing news articles and commenting on them. In this paper, we propose an approach for identifying “comment-tweets” that comment on news articles. We discuss the nature of comment-tweets and compare them to subjective tweets. We utilize a machine learning-based classification approach for distinguishing between comment-tweets and others that only report the news. Our approach is evaluated on the TREC-2011 Microblog track data after applying additional annotations to tweets containing comments. Results show the effectiveness of our classification approach. Furthermore, we demonstrate the effectiveness of our approach on live news articles.