Citation Context

...ent analysis has focused on understanding the sentiments of individual tweets [2, 11, 6, 31, 28, 15], although more recently some authors have addressed the problem of inferring user-level sentiments =-=[29, 17]-=-; see Section 6 for more details. Smith et al. [27] and Deng et al. [7] study both tweet-level and user-level sentiments, but they assume that a user’s sentiment can be estimated by aggregating the se...

Citation Context

... user’s overall sentiment toward a target, or predicting a user’s opinion/sentiment toward a topic/target. Kim et al. described user-level sentiment prediction using collaborative filtering approach (=-=Kim et al. 2013-=-). Tan et al. described user-level sentiment analysis using social network information (Tan et al. 2011). However, none of the existing research on opinion and sentiment analysis predicts the likeliho...

Citation Context

...ations and fields of knowledge, such as health education (Luque et al.,s2009), consumer reviews (e.g., Epinions.com) (Massa & Avesani, 2007), and sentimentsprediction in twitter conversation threads (=-=Kim et al., 2013-=-). New approaches have alsosbeen developed, like the one introduced by Cai et al. (2014), who propose a user-basedsrecommendation that makes a representation of the user through a vector that cansindi...

Citation Context

... dynamics. For example, political parties routinely use social media to sense people’s opinion about their political discourse1; quantitative investment firms measure investor sentiment and trade using social media [18]; and, corporations leverage brand sentiment, estimated from users’ posts, likes and shares in social media and social networking sites, to design their marketing campaigns2. In this context, multiple methods for sensing opinions, typically based on sentiment analysis [21], have been proposed in recent years. However, methods for accurately forecasting opinions are still scarce [7, 8, 19], despite the extensive literature on theoretical models of opinion dynamics [6, 9]. In this paper, we develop a novel modeling framework of opinion dynamics in social media and social networking sites, SLANT3, which allows for accurate forecasting of individual users’ opinions. The proposed framework is based on two simple intuitive ideas: i) users’ opinions are hidden until they decide to share it with their friends (or neighbors); and, ii) users may update their opinions about a particular topic by learning from the opinions shared by their friends. While the latter is one of the main under...

Citation Context

... infer user’s sentiment toward a topic in social media or online forums. Tan et al. [26] proposed a semi-supervised approach to inferring users’ sentiment using social network information. Kim et al. =-=[12]-=- and Gao et al. [8] proposed to use a collaborative filtering like approach to estimate user-level sentiment. Lu et al. [17] proposed to use content and social interactions to discover opinion network...