ULM: A user-level model for emotion prediction in social networks

Abstract

Users express their emotions in online social networks (OSNs). Studying emotions is important for understanding user behaviors. Existing methods in emotion prediction mainly use personal emotion and friend emotion to predict target emotion. In this study, public sentiment is introduced to denote the sentiment of the majority in the network. Public conformity is calculated to measure the degree of a user conforming to the public sentiment. According to the public conformity, users are categorized into three classes: Approvers, independents, and starters. A user-level model for emotion prediction is proposed to predict target emotions of different classes of users, taking into account of the public sentiment, individual sentiment, friend sentiment and pseudo-friend sentiment. Relations between the public conformity and the structure of the network are studied. Experiments conducted on Sina Weibo show that the proposed model could achieve performance improvements to some existing methods in most cases.