| Summary: | 碩士 === 國立政治大學 === 資訊科學學系 === 105 === When a new event happens, related messages can not only be released by the press but also through various network channels by the general public. In recent years, social media have become a popular and important platform for news reporters and the general public to propagate news messages. However, due to the massive amount of messages on the social media platforms, it is very time consuming to manually collect these data in order to grasp the development of a news event over time. In this research, we aim to develop a visualization system that can help researchers find key tweets on twitters for a news event in an incremental manner. Our system allows a user to customize design parameters for finding key tweets from various aspects in order to understand how a news event evolve over time on social media. We have invited four participants to test use our system through a tutorial, guided tasks, and free exploration. A questionnaire survey and interview were conducted after the experiments. The evaluation results show that the average score for the usability dimension is 4.1 out of 5, showing that the system can effectively assist the users in analyzing the development of a news event. The average score for the ease-of-use dimension is 4.3, meaning that most users agree that our system is easy to use. The results reveal that our research goals of helping users find key tweets and understand news development have been achieved and the development of such a visualization system is valuable for news analysis on social media.
|
|---|
