[1J3] [General Session] 9. NLP / IR

[1J3-04] Preliminary Analysis of Weibo Emojis for Sentiment Analysis of Chinese Social Media

〇Da Li¹, Rafal Rzepka¹, Kenji Araki¹ (1. Hokkaido University)

Keywords:Sentiment Analysis, Emoticons, Weibo

In this paper, we introduce our preliminary analysis of emoticons used on Weibo, Chinese microblog. By performing a polarity annotation with a new humorous type added, we have confirmed that 23 emoticons can be considered more as humorous than positive or negative. We also discussed some possible related problems which might occur during any social media sentiment analysis performed on Chinese language. We believe that combining machine learning, sentiment lexicons and paralinguistic elements as emoticons can be expected to achieve better performance in the future.