| Summary: | 碩士 === 國立政治大學 === 資訊科學系碩士在職專班 === 104 === In recent years, Facebook has become a very popular social media with millions of users. Besides personal walls, many celebrities and organizations leverage the public Facebook pages to engage their fans for developing their brands or promoting their ideas. Due to their public nature, Facebook pages have become good targets of social communication research by scholars.
In this thesis, we propose to investigate the relations among Facebook pages built by post sharing activity during public events. Toward this goal, we have designed and implemented a system for analyzing posts sharing among Facebook pages. It allows users to set initial seed pages, related parameters and duration of collection on the public events they concern. Our system will collect specified pages data (posts, comments, likes) of original seed pages through Facebook graph API. After the data collection period, the system will analyze whether the new pages fit the user specified criteria, and adding the new pages into the page pool for next stage of collection. In other words, our system will iteratively collect pages related by post sharing.
As an evaluation of our system, we conducted an experiment based on the Sunflower Movement and use Facebook posts of seed pages to find new similar pages to construct groups of similar pages. The experimental results show that our system can collect many related pages and help scholars to group them by sharing in a meaningful way.
|
|---|
