A Study of Travel Notes on Blogs based on Text Mining and Machine Learning

• Main Authors: Yeh, Shan-Yu, 葉珊妤
• Other Authors: Li-Ching Ma
• Format: Others
• Language: zh-TW
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/vee897

碩士 === 國立聯合大學 === 資訊管理學系碩士班 === 106 === Modern people are often under a variety of pressures. Traveling is a good and popular way for people to relieve their stresses. People usually collect information and plan their trips before setting out. Travel notes on blog are often regarded as important references. This study takes travel notes on blogs regarding Miaoli County as an example to analyze the relationships among attractions, restaurants and accommodations based on association rules, and presents rich information through simple graphics. In addition, a text mining approach is employed to analyze the contents and critical terms of the travel notes. Three machine learning methods including support vector machine, decision tree and backpropagation neural network, and one deep learning approach, deep neural network, are adopted to predict the amount of views and comments of travel notes on blogs. The results of this study can not only be used as references for tourists to plan their trips but also provide valuable information for bloggers to write more attractive travel notes to catch more peoples’ attentions.