Fuzzy rules reduction using rough set approach
This paper presents the use of Rough Set approach to compute reducts and generate concise fuzzy rules from a fuzzy rule base system of a student model. The purpose of modeling the student is to evaluate the students conceptual understanding (i.e. performance level and learning efficiency) in learni...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Conference or Workshop Item |
| Published: |
2003
|
| Subjects: | |
| Online Access: | http://eprints.utm.my/id/eprint/3399/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This paper presents the use of Rough Set approach to compute reducts and generate concise fuzzy rules from a fuzzy rule base system of a student model. The purpose of modeling the student is to evaluate the students conceptual understanding (i.e. performance level and learning efficiency) in learning C programming language. Based on the Rough Set approach, the fuzzy rule base system that consists of four antecedents and two consequents is transformed into a decision table with four conditional attributes and a single decision attribute. Johnson reducer and Genetic Algorithm reducer are the methods used for computing reducts. Experimental results have shown that Rough Set approach has successfully reduced the fuzzy rules optimally. The number of rules being reduced depends on the refinement and the consistencies of the decision attribute values. After comparing the defuzzified values of the original fuzzy rule base system with the reduced fuzzy rule base system, no obvious degradation of performance occurred. Thus, the reduced fuzzy rule base is said to preserve the same performance as the original fuzzy rule base system. |
|---|