Categorized tip information of social media based on topic detection
A. Nadamoto & Y. Hattori · 2014
Usually the social media community is based on some theme. However, in the community, there are many kinds of topics in a social media community, because users write contents freely. If the theme of social media community is “fireworks festival”, there might be a traffic topic, restaurant topic, best seat topic, or a children topic, each of which might include the theme. Then we detect the topics from the community and cluster the comment based on the topics. Subsequently, we extract tip information from each cluster. Many methods can be used to detect the topic. Tsutsumida (K. Tsutsumida & Uchiyama 2012) proposed that, when detecting the topic from sparse data, the Latent Dirichlet Allocation (LDA) (D.M. Blei 2003), (K. Dave & Pennock 2003b) and graphbased methods are better than the other methods. As described in this paper, we use LDA methods to detect the topic from the social media content, which is sparse data.