New Query Expansion Methods Based on Fuzzy Rules and Genetic Algorithms
碩士 === 國立臺灣科技大學 === 資訊工程系 === 92 === In recent years, many methods have been proposed to deal with document retrieval using soft computing techniques, where the most popular way is to analyze the information of relevant documents for modifying the user’s query to increase the performance of document...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | en_US |
| Published: |
2004
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/03899146591845570641 |
| Summary: | 碩士 === 國立臺灣科技大學 === 資訊工程系 === 92 === In recent years, many methods have been proposed to deal with document retrieval using soft computing techniques, where the most popular way is to analyze the information of relevant documents for modifying the user’s query to increase the performance of document retrieval systems based on the user’s relevance feedback. In this thesis, we propose two new methods to modify and to reweight the user’s query based on fuzzy rules and genetic algorithms. The first method infers the relevant degrees of relevant query terms and then expands the user’s query based on fuzzy rules. The second method searches the best weights of query terms of user’s query vectors based on genetic algorithms to increase the recall rate and the precision rate for document retrieval. The proposed methods can improve the performance of document retrieval systems.
|
|---|