Integrating rough set and genetic algorithm for negative rule extraction

Authors:
Junyu Liu;Yubao Liu;Yan Long
Affiliations:
Department of Computer Science, Sun Yat-sen University, Guangzhou, China;Department of Computer Science, Sun Yat-sen University, Guangzhou, China;Department of Computer Science, Sun Yat-sen University, Guangzhou, China
Venue:
IDEAL'09 Proceedings of the 10th international conference on Intelligent data engineering and automated learning
Year:
2009

Citing 6
Cited 0

C4.5: programs for machine learning

C4.5: programs for machine learning
Mining association rules between sets of items in large databases

SIGMOD '93 Proceedings of the 1993 ACM SIGMOD international conference on Management of data
The Power of Decision Tables

ECML '95 Proceedings of the 8th European Conference on Machine Learning
Mining diagnostic rules from clinical databases using rough sets and medical diagnostic model

Information Sciences: an International Journal - Special issue: Medical expert systems
Mining positive and negative association rules: an approach for confined rules

PKDD '04 Proceedings of the 8th European Conference on Principles and Practice of Knowledge Discovery in Databases
Data Mining: Practical Machine Learning Tools and Techniques, Second Edition (Morgan Kaufmann Series in Data Management Systems)

Data Mining: Practical Machine Learning Tools and Techniques, Second Edition (Morgan Kaufmann Series in Data Management Systems)

Quantified Score

Hi-index	0.00

Visualization

Abstract

Rule extraction is an important issue in data mining field. In this paper, we study the extraction problem for the complete negative rules of the form ¬R → ¬D. By integrating rough set theory and genetic algorithm, we propose a coverage matrix based on rough set to interpret the solution space and then transform the negative rule extraction into set cover problem which can be solved by genetic algorithm. We also develop a rule extraction system based on the existing data mining platform. Finally, we compare our approach with other related approaches in terms of F measure. The comparison experimental results on the real medical and benchmark datasets show that our approach performs efficiently for incompatible and value missing data.