The lexicon-based sentiment analysis for fan page ranking in Facebook
The traditional methods to rank a Facebook fan page only rely on the user engagement including the number of posts, comments, and “likes”. The polarity of each comment, which can be positive, neutral, or negative, is ignored in these methods. In this paper, we propose a content-based ranking method in which the user engagement and the comment polarity are all considered. The user comment is analyzed using a lexicon-based approach. We apply the proposed method for the real Facebook dataset collected using the Social Packets crawler. The result shows that the ranks of pages estimated by our method is close to the ranks estimated by engagement based method. More importantly, by concerning the comment polarity, our page ranking is more accurate regarding user opinion.
Abstract: The traditional methods to rank a Facebook fan page only rely on the user engagement including the number of posts, comments, and “likes”.