| Summary: | 碩士 === 國立清華大學 === 資訊系統與應用研究所 === 105 === With the evolution of Internet technologies, especially in Web applications, news articles flood the Web every day from an extreme amount of news portals from around the world. It is almost impossible for a single person to keep track of an event, or a series of related events. Based on this problem, news services and search engines have become more popular. Despite of the ability to organize the news based on content similarity or categories, the result can't represent user perspective. By all means, the main objective of this research is to propose a different news articles clustering method based on users' perspective by using sentiment analysis and compare it to traditional clustering method. After the experiments, we could see that clustering the news articles with sentiment analysis had a higher similarity according to users' real world usage in comparison to the traditional clustering. Hence, clustering news articles based on sentiment analysis is indeed a viable way to match users' perspective.
|
|---|
