Title: The classification and visualization of trending topics in online word-of-mouth data
Authors: Atsuho Nakayama - Tokyo Metropolitan University (Japan) [presenting]
Abstract: Trending topics in online word-of-mouth data are classified by focusing on topics related to new products. The analysis of large amounts of online word-of-mouth data collected from Social Networking Service has received much attention to help identify market trends. Twitter has been widely using in Japan. Consumers post a lot of comments regarding a wide range of topics in Twitter. The tweets include opinions about products and services. We collected Twitter entries about new products based on their specific expressions of sentiment. We tokenized each tweet message that was written in sentences or sets of words to detect topics more easily. Morphological analyses has been that such as tokenization, stemming, and part-of-speech tagging to separate the words. Next, we selected keywords representative of our chosen topics. We performed a statistical analysis based on the complementary similarity measure that has been widely applied in the area of character recognition. Then, we detected trending topics related to a new product by classifying words into clusters based on the co-occurrence of words in Twitter entries. Topic-based sentiment analysis has been used to extract and visualize the topic of customer interests and opinions.