Ontological Inference for User Intention Extraction、Query Expansion and Concept-based Retrieval

• Main Authors: Yi-Hsuan Hung, 洪奕璿
• Other Authors: Shiow-Yang Wu
• Format: Others
• Language: zh-TW
• Published: 2004
• Online Access: http://ndltd.ncl.edu.tw/handle/22279160590058629648

碩士 === 國立東華大學 === 資訊工程學系 === 92 === The using of keyword-based retrieval is always easy and convenient for end users; however, the result hardly reaches the goal of precisely revealing the concept in document. Adopting concept-based retrieval improves the limitations in using keyword-based. The result of concept-based retrieval meets directly with the user intention in searching instead of merely the words. Therefore, in order to demonstrate the advantages in concept-based, it is very important for extract out the user intention of the query. In the thesis, we bring out the usage of adopting the user intention extraction, query expansion, and concept-based retrieval in ontology inference. The method helps in inferring the relationship among concepts, and avoids the ambiguous in blurred concepts. The user focus tree designed for users displays the conceptual relationship during the searching, and it was also used in query expansion and concept-based retrieval. The extracted document will first calculate the relative degree within users'' intention, and the user focus tree then becomes the base for calculating various weights. We can show the most related information for users with it. By using these methods, users no longer have to worry about depiction on their concept or the extended one during the searching. In light of this, more conceptual-related information with accuracy will be found successfully by adopting ontology inference and user intention extraction during the searching.