A Study on Adding New Concepts to Domain Ontology

• Main Authors: Ya-Ching Lee, 李雅清
• Other Authors: Rung-Ching Chen
• Format: Others
• Language: en_US
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/4gn4ng

碩士 === 朝陽科技大學 === 資訊管理系碩士班 === 94 === In recent years, ontology application has become one of the most popular research topics on the Internet. Many researchers use domain ontology to facilitate knowledge sharing and reusing. However, the specification of the ontology is still incompletely now; due to no standard expression and creating method exist for the ontology. The information is changed fast on the Internet, so the knowledge on the ontology often can not really meet the user’s requirement. Therefore, how to create a renewable ontology is necessary. In this paper, we will propose an ontology renewable method which can insert new keywords into the corresponding constructed ontology. The novel method uses TF-IDF (Term Frequency – Inverse Document Frequency) and LSA (Latent Semantic Analysis) to strengthen the semantic characteristic of keywords and transforms the LSA matrix to a high dimensional space based on collected web page. The similar values between keywords and concepts are gotten by comparing the high dimension keyword to the corresponding constructed ontology concepts based on group center similarity. The keyword with the least value of similarity will be inserted into the offspring of the concept on the domain ontology and become the domain concept. The primary experiment results indicate the method has 73% accuracy.