Term-based Knowledge Mining and Editing

• Main Authors: Chun-Hao Lei, 雷鈞皓
• Other Authors: Sheng Wu
• Format: Others
• Language: zh-TW
• Online Access: http://ndltd.ncl.edu.tw/handle/60137555827308382270

碩士 === 國立中正大學 === 資訊工程所 === 96 === Knowledge base is an important part in many researches. There are two ways to construct a knowledge base, the knowledge mining and knowledge editing. However, there are very few knowledge base systems that can combine both approaches together. As a result, it''s hard to build a knowledge base that is very comprehensive and is of great quality. In this thesis, we improve the functions of Sopedia system and introduce some new designs to build a new knowledge editing system called NUPedia. Furthermore, we analyze a large corpus of Web data, mining the term relations and interesting terms in the corpus, develop ways to identify the key terms of a document, and combine these approaches into the NUPedia system to make it a term-based knowledge base system.