Topic Knowledge Structure for Dentistry Test Bank

• Main Authors: PO-CHIN LI, 李柏辰
• Other Authors: Deng-Yiv Chiu
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/10148280286657705824

碩士 === 中華大學 === 資訊管理學系 === 104 === Users can search and get huge information from search engines today. However, the amount of information that we get is extremely huge. It is hard and time-consuming for users to find necessary knowledge. Therefore, how to assist users to obtain necessary information from the Internet and to retrieve implicit knowledge contained in the internet is a major goal of knowledge management. The national examinations evaluate the professional knowledge to select suitable candidates. Therefore, it is very important to explore the representativeness and property of the national examinations. In this study, we construct an appropriate knowledge structure to analyze the dentistry test bank devised by national exams institution. We use CKIP Chinese tokenization system developed by Taiwan Academia Sinica to separate the sentences into terms. We use Information Retrieval （IR）and Vector Space Model （VSM） to present the test questions and features. The relatedness between topics and features is computed. Independency of topic is further calculated via independent chi-square test. According to the fitness value, the representativeness of the selected topics can be determined. The genetic algorithm would be performed to construct an appropriate knowledge structure. In our experiment, topic knowledge structure for dentistry test bank can not only effectively present representative topics, related key words and issues, but also help understand the research tendencies by listing examinations in descending according to the degree of correlation related questions. Therefore, it is very important to analyze suitable and representativeness for dentistry test bank.