Bachelor and Master Theses

Title: Generating Fuzzy Rules from Case Base for Classification Problems
Subject: Computer Science
Level: Advanced
Description: Case-based reasoning is a good technique to solve new problems based on previous successful cases; CBR shows significant promise for improving the effectiveness of complex and unstructured decision-making problems. The main assumption of CBR is that similar problems have similar solutions. But the concept of similar can vary in different situations and remain an issue. Utility oriented similar modeling has been identified as a significant direction for Case-based reasoning research.
So far the main stream of works is focused on similarity modeling, using similarity to represent utility. In the thesis we propose a new way to represent the case utility rather than use similarity modeling, because the similarity model may not contain all of information about the cases, and we use modified WANG’s algorithm which derives from WANG’s method to generate fuzzy if-then rules from case library to represent the case utility, because we believe the fuzzy if-then rules present more powerful and flexible means to capture domain information for case utility than traditional similarity measures based on feature weighting. Fuzzy rule-based reasoning is utilized as a case matching mechanism to determine whether and to which extent a known case in case library is useful to a new problem. The reason why we choose the WANG’s algorithm is because WANG’s algorithm is a simpler and faster algorithm to generate if-then rules. The way we generate fuzzy if-then rule is by using case-pairs with solutions in case library. The experiment on several data sets has showed superiority of our work over traditional schemes as well as the feasibility of learning fuzzy if-then rules from a small number case base while still having good performances.
Prel. end date: 2012-06-14
Presentation date: 2012-06-14
Student: Liangjun Ma lma11001@student.mdh.se
Student: Shouchuan Zhang szg11001@student.mdh.se
IDT supervisor: Ning Xiong
ning.xiong@mdh.se, +46-21-151716

Rapport och bilagor

Size

Senaste uppdatering

TR1387.doc

948736

2012-12-07, 00:52

TR1387.pdf

1115519

2012-12-07, 00:53


  • Mälardalen University |
  • Box 883 |
  • 721 23 Västerås/Eskilstuna |
  • 021-101300, 016-153600 |
  • webmaster |
  • Latest update: 2018.03.15