以社交網路為基礎的個人知識收藏推薦系統

• Main Authors: Chung-chieh Hsia, 夏中傑
• Other Authors: Cheng-lung Huang
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/29436149853918847346

碩士 === 國立高雄第一科技大學 === 資訊管理系企業電子化碩士班 === 104 === The social bookmarking sharing system is a powerful platform for people to read conveniently, store easily, and select fast the information on the internet. In the bookmarking sharing systems, providing preference information for users is an important issue. From the social networks users can explore internet resoureces that they are inerested in. This study calculates the relationship strength between the target user and the target user''s followings and followers to find top-N similar neighbors. From the top-N similar neighbors, the articles (internet resources) that may interest the target user were predicted an recommended to the target user. This study collected the experimental dataset from Diigo, which is a famous bookmarking sharing system. The experimental results show that the proposed recommendation model is more accurate than the Random model and Recency model. Although the proposed model was not as good as the cosine model in accuacy, however, in the metrics of the diversity and executing time, the proposed model were better than the cosine model. The model which this research recommends filters a great amount of useless information efficiently and provides users with relative and diverse information in short time. Users only need to spend a little time to acknowledge about what their friends recently followed and it''s what you are interested in.