Classification rule discovery with ant colony optimization

Ant-based algorithms or ant colony optimization (ACO) algorithms have been applied successfully to combinatorial optimization problems. More recently, Parpinelli and colleagues applied ACO to data mining classification problems, where they introduced a classification algorithm called Ant/spl I.bar/Miner. In this paper, we present an improvement to Ant/spl I.bar/Miner (we call it Ant/spl I.bar/Miner3). The proposed version was tested on two standard problems and performed better than the original Ant/spl I.bar/Miner algorithm.

[1]  Hussein A. Abbass,et al.  Introducing Data Mining and Knowledge Discovery , 2002 .

[2]  Marco Dorigo,et al.  Ant system: optimization by a colony of cooperating agents , 1996, IEEE Trans. Syst. Man Cybern. Part B.

[3]  Alex A. Freitas,et al.  An Ant Colony Algorithm for Classification Rule Discovery , 2002 .

[4]  Barbara Webb,et al.  Swarm Intelligence: From Natural to Artificial Systems , 2002, Connect. Sci..

[5]  Luca Maria Gambardella,et al.  Ant Algorithms for Discrete Optimization , 1999, Artificial Life.

[6]  B. Naudts,et al.  Ant colonies are good at solving constraint satisfaction problems , 2000, Proceedings of the 2000 Congress on Evolutionary Computation. CEC00 (Cat. No.00TH8512).

[7]  Ruoying Sun,et al.  Multiagent reinforcement learning method with an improved ant colony system , 2001, 2001 IEEE International Conference on Systems, Man and Cybernetics. e-Systems and e-Man for Cybernetics in Cyberspace (Cat.No.01CH37236).

[8]  Bo Liu,et al.  Density-Based Heuristic for Rule Discovery with Ant-Miner , 2002 .