초록 열기/닫기 버튼
Recently, many users have shared their experiences and evaluations of various leisure activities using rating systems. For instance, movie information sites present users' reviews and ratings on various movies, and then other users employ these reviews and ratings to select movies to watch. However, these traditional rating systems suffer from some limitations, including the influence of the subjectivity of the participants on the ratings, because the ratings are usually calculated merely by considering the average score of all participants. Therefore, the ratings of leisure products and services are determined not only by their original characteristics but also by who participated in the rating events. This phenomenon inevitably impairs the reliability of rating systems. Therefore, we propose a methodology for improving the reliability of rating systems by moderating personal factors from the raw ratings. This study then analyzes the possibilities for the practical use of the proposed methodology through its application to 5,000 reviews and ratings for movies acquired from the IMDb site.
키워드열기/닫기 버튼
Leisure Activity, Movie Industry, Text Mining, Rating System
