Semi-supervised microblog sentiment analysis using social relation and text similarity

Microblog Sentiment Analysis (MSA) is a popular and important theme in social networks. Microblog platform such as Twitter, can collect rich microblogging messages everyday. However, for MSA tasks, it is still difficult and costly to collect sufficient manual sentiment labels for training. There are rich unlabeled microblogging messages, but only a few manual labeled messages. In this paper, we propose a novel semi-supervised learning approach for MSA. Specifically, we make use of microblog-microblog relations to build a graph-based semi-supervised classifier. We incorporate social relations and text similarities into building microblog-microblog relations. Our model connects labeled data and unlabeled data via microblog-microblog relations. Experiments on two real-world datasets show that our graph-based semi-supervised model outperforms the existing state-of-the-art models.