A Study on Personalized Knowledge Recommendation Mechanism by Applying Web Mining Technology：A Study of Online Travel Blogs

• Main Authors: Ya-Chien Lu, 呂亞蒨
• Other Authors: Wen-Shan Lin
• Format: Others
• Language: zh-TW
• Online Access: http://ndltd.ncl.edu.tw/handle/76650220596539161805

碩士 === 國立嘉義大學 === 資訊管理學系碩士班 === 96 === As the network technology and equipments to the popularization of the Internet, it becomes a major information and knowledge source for people. However, there is a need to improve the information overload problem. Take the recommendation systems as the cases, the adopted recommendation mechanisms can be categorized as content-oriented, collaborative and knowledge-based filtering ones. Each mechanism has its pros and cons. The knowledge-oriented hybrid recommendation mechanisms is denoted as the most efficient one in assisting online information search. In this study, the case of online blog is taken for further examination due to the fact that it has become a phenomenon in the Internet. People take delight in surfing blogs. However, online blogs fall in unstandardized formats. It makes the recommendation system need to solve the problems of natural language usage problem. Therefore, the idea of adopting knowledge-based information recommendation mechanism combined with web mining technology is set to be tested experimentally in this study. This thesis aims at investigating the notion of recommendation system from the literature and test the effectiveness of hybrid recommendation mechanism by a user-centered method. The experimental method is adopted. Results indicate that the knowledge-based hybrid recommendation mechanism with the combinations of content-based and collaborative filtering mechanisms outperform the others statistically significant. Results and future research opportunities are discussed and given. This study is served as the basis for the improvements of recommendation systems and the related areas.