Hybrid Ant Colony Optimization and Simulated Annealing for Rule Induction

This paper proposes a hybrid of ant colony optimization and simulated annealing for rule induction. The hybrid algorithm is part of the sequential covering algorithm which is the commonly used algorithm to extract classification rules directly from data. The hybrid algorithm will minimize the proble...

Full description

Saved in:
Bibliographic Details
Main Authors: Saian, R., Ku-Mahamud, Ku Ruhana
Format: Conference Proceeding
Language:English
Subjects:
Accuracy
Ant colony optimization
Ant-Miner
Breast cancer
classification
Classification algorithms
Data mining
Prediction algorithms
Rule induction
Simulated annealing
Training data
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper proposes a hybrid of ant colony optimization and simulated annealing for rule induction. The hybrid algorithm is part of the sequential covering algorithm which is the commonly used algorithm to extract classification rules directly from data. The hybrid algorithm will minimize the problem of low quality discovered rule by an ant in a colony, where the rule discovered by an ant is not the best quality rule. Simulated Annealing will be used to produce a rule for each ant. The best rule for a colony will then be chosen and later the best rule among the colonies will be included in the rule set. The ordered rule set is arranged in decreasing order of generation. Thirteen data sets which consist of discrete and continuous data from UCI repository were used to evaluate the performance of the proposed algorithm. Promising results were obtained when compared to the Ant-Miner algorithm in terms of accuracy, number of rules and number of terms in the rules.
DOI:10.1109/EMS.2011.17