A Study on Web-Based Bookmark System --Applying Ontological and Collaborative Filtering Approaches for Teaching Material Sharing at Compulsory Education

• Main Authors: William Lee, 李武霖
• Other Authors: Jin-Tan David Yang
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/47383498820529876661

碩士 === 國立高雄師範大學 === 資訊教育研究所 === 93 === The purpose of this study is to implement a web bookmark system for professional community of teachers at compulsory education in Taiwan. As those teachers are looking for teaching materials on the web, they feel an extra burden beyond daily instruction tasks. They can hardly locate what materials they want within a short time. however, demand for community bookmarks is getting stronger while the internet has offered many useful teaching materials for education. This study proposes combining ontological and collaborative filtering approaches to recommend web-based teaching materials or share domain ontologies among professional teachers’ community. The recommendation is based on Pearson correlation matrix. The approach is collective intelligence to enhance teaching materials’ sharing. This methodology of this study consists of three stages. Firstly, the Google searching engine is used as a web service interface. Secondly, domain ontologies of nine-year curriculum is constructed in web-based bookmark system and visually presented knowledge maps. Thirdly, a collaborative filtering algorithm provides teachers with locating web teaching materials on the dynamic nodes. The collaborative filtering is included for recommending teaching materials or ontologies duplication. Additionally, in our study, we utilize the SCTNET's(http://sctnet.edu.tw) teaching material share as basis of emulation data by randomized bookmarks for each teachers. Two primary parameters used in this study are threshold for peer annotations and Pearson correlation among teachers. The optimal recall and precision value for these two parameters in terms of bookmark thresholds and Pearson correlation are 3 and 0.75. The contributions of this paper are as follows: (a) This study extends the traditional bookmark into web-based edition; (b) Web-based bookmark enhances teaching materials’ sharing among professional community; (c) Development of collaborative filtering provides reliable recommendations on ontologies or teaching materials; (d). Knowledge map is visually represented for those teachers to understand the whole knowledge picture. Finally, the implications of this study are also included.