Classification rule discovery with ant colony optimization algorithm

Authors:
Samaira Hodnefjell;Ilaim Costa Junior
Affiliations:
Computer Science Department, Federal University of Juiz de Fora, Brazil;Institute of Computing, Computer Science Department, Fluminense Federal University, Brazil
Venue:
IDEAL'12 Proceedings of the 13th international conference on Intelligent Data Engineering and Automated Learning
Year:
2012

Citing 8
Cited 0

Elements of information theory

Elements of information theory
C4.5: programs for machine learning

C4.5: programs for machine learning
The ant colony optimization meta-heuristic

New ideas in optimization
Ant algorithms for discrete optimization

Artificial Life
Data Mining: An Overview from a Database Perspective

IEEE Transactions on Knowledge and Data Engineering
The CN2 Induction Algorithm

Machine Learning
The Fuzzy Ant

IEEE Computational Intelligence Magazine
Data mining with an ant colony optimization algorithm

IEEE Transactions on Evolutionary Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

Ant Colony Algorithms has been successfully applied to solve combinatorial optimization problems. Subsequently applications on Data Mining (DM) appeared, more specifically aiming to solve classification problems. The Ant-Miner [3] algorithm is a good example of a solution to this problem. This algorithm is better than C4.5 [7] and CN2 [8]. This paper presents a new algorithm which applies an innovative modeling of the foraging behavior of ants [4] to the Ant-Miner. As a result of this adaptation, four different versions of the Ant-Miner algorithm were generated, tested and compared to the original version using seven public domain data sets. One of the versions produced comparatively superior results in terms of predictive accuracy in different system configurations.