Opinion Classification in Professional Sports Fan Sites using Topic Keyword-Based Sentiment Analysis

Hyungho Byun, Sihyun Jeong, Chong-kwon Kim

http://doi.org/10.5626/JOK.2018.45.4.390

In this study, we propose the classification method using topic keyword-based sentiment analysis through the posts of professional sports fan sites in Korea. We studied ways to take into account the use of special communication methods or vocabulary in the community and defined keywords based on the characteristics of the topic or frequency of the community"s words. In addition, we presented a new sentiment analysis approach that utilizes the use of keyword pools and the proximity relation to keywords. Through three years of actual community dataset, sentiment analysis based on the topic keyword is more effective than the existing method and reflects the community environment.