Using Text Mining to Extract Issues for School:An Empirical Study of The Social Platform-Dcard

碩士 === 國立臺中科技大學 === 資訊管理系碩士班 === 106 === Sentiment analysis has become one of the most popular of the text exploration in the word. Many social platforms have become the most popular text exploration and analysis platforms, such as Facebook, Twitter, Instagram, etc. "Dcard" is the largest...

Full description

Bibliographic Details
Main Authors: HSIN-YI WANG, 王馨憶
Other Authors: JUNG-WEN LO
Format: Others
Language:zh-TW
Published: 2018
Online Access:http://ndltd.ncl.edu.tw/handle/h75ntt
Description
Summary:碩士 === 國立臺中科技大學 === 資訊管理系碩士班 === 106 === Sentiment analysis has become one of the most popular of the text exploration in the word. Many social platforms have become the most popular text exploration and analysis platforms, such as Facebook, Twitter, Instagram, etc. "Dcard" is the largest social networking platform for college students in Taiwan. It is anonymous on the platform, so users are mostly willing to freely post their ideas and opinions on the platform. Since the users of the Dcard are college students in Taiwan, there are many articles related to the school. We have executed sentiment analysis to the articles on the Dcard, so the school can refer to the information for improvement. The Python language is used in this study to obtain meaningful information from the Dcard platform. In order to achieve this purpose, we have added keywords to expand the vocabulary in our sentiment dictionary. We added 32 words of lexical vocabulary, 21 positive words and 10 negative words in sentiment dictionary, and new phonetic symbols such as English letters, face text in stop words. These additional effort makes the research more accurate when we perform the analysis.