WeiboFinder: A Topic-Based Chinese Word Finding and Learning System.

Wen-hao Chen,Yi Cai,Kin Keung Lai,Li Yao,Jun Zhang,Jingjing Li,Xingdong Jia

DOI: https://doi.org/10.1007/978-3-319-66733-1_4

2017-01-01

Abstract:With the explosive growth of user-generated data in social media websites such as Twitter and Weibo, a lot of research has been conducted on using user-generated data for web-based learning. Finding users’ desired data in an effective way is critical for language learners. Social media websites provide diversified data for language learners and some new words such as cyberspeak could only be learned in these online communities. In this paper, we present a system called WeiboFinder to suggest topic-based words and documents related to a target word for Chinese learners. All the words and documents are from the Chinese social media website: Weibo. Weibo is one of the largest microblog social meida websites in China which has similar functions as Twitter. The experimental results show that the proposed method is effective and better than other methods. The topics from our method are more interpretable and topic-based words are useful for Chinese learners.