An emotion-based independent cascade model for sentiment spreading

Online social networks (OSNs) provide a platform for users to publish messages, by which users express their emotions on events or products. The phenomenon that emotions are spread by retweeting messages is referred to as sentiment spreading. In this paper, an emotion-based independent cascade model is proposed to study the process of sentiment spreading. The proposed model divides the process of sentiment spreading into three steps. First, propagation probabilities are introduced to predict whether users retweet messages. Second, a learning model taking account of user features, structural features, and tweet features is applied to predict whether emotions are changed after retweeting. Third, the transforming weights are calculated to predict what the sentiments of the retweets transform to. The experimental results on Sina Weibo demonstrated that the proposed model could achieve 15.78% and 4.9% performance improvements compared with two baseline methods.