IOSR Journal of Computer Engineering (IOSR-JCE) e-ISSN: 2278-0661, p- ISSN: 2278-8727Volume 16, Issue 3, Ver. II (May-Jun. 2014), PP 63-72 www.iosrjournals.org www.iosrjournals.org 63 | Page Enhanced cAntMiner PB Algorithm for Induction of Classification Rules using Ant Colony Approach Safeya Rajpiplawala 1 , Dheeraj Kumar Singh 2 1 (Computer Science and Engineering, Parul Institute of Engineering and Technology/ Gujarat Technological University, India) 2 (Information Technology, Parul Institute of Engineering and Technology/ Gujarat Technological University, India) Abstract : Mining classification rules from data is a key mission of data mining and is getting great attention in recent years. Rule induction is a method used in data mining where the desired output is a set of Rules or Statements that characterize the data. Within Rule Induction model, Swarm Intelligence (SI) is a technique where rules may be discovered through the study of joint behavior in decentralized, self-organized systems, such as ants. Ant-Miner is a rule induction algorithm that uses SI techniques to form rules. The main idea of this study is to discover the suitability of ant colony optimization for constructing accurate classifiers which can be learned in practical time even for big datasets. The cAntMiner PB algorithm is an extension of the cAntMiner algorithm. The main task is to modify the existing algorithm cAntMiner PB to allow each rule to dynamically select rule quality evaluation function and to improve the accuracy and preserving rule list simplicity. In this study, we examine the use of different rule quality evaluation functions for rule quality assessment prior to pheromone update and check how the use of different evaluation function affects the output model in terms of predictive accuracy and model size. In experimental results, we use 10 different rule quality evaluation functions on 12 benchmark datasets and found that predictive accuracy obtained by new proposed method is statistically significantly higher than the predictive accuracy of existing algorithm. Keywords: ACO, SI, Rule Classifier, AntMiner I. INTRODUCTION There is an extensive variety of algorithms that have come out from the actions of social bugs. Social insects are typically differentiated by their self union and with the negligible contact or the absence of it. Every social insect separately is superior. They can get information regarding surroundings and cooperate with the remote insects in some way, by stigmergy. All these features describe Swarm Intelligence. In section 2 details are described of ACO and AntMiner Algorithm, in section 3 cAntMiner Algorithm, in section 4 cAntMiner PB Algorithm, in section 5 proposed work, in section 6 implementation tools and in section 7 Experimental results are shown on 12 datasets taken from UCI Machine Learning Repository. Finally in section 8 Applications and in Section 9 Conclusion is described. II. ACO AND ANTMINER ALGORITHM Ant colony optimization (ACO) [1] is a stem of a recently developed form of artificial intelligence known as swarm intelligence. Swarm intelligence (SI) is “the property of a structure whereby the combined activities of (simple) agents work together locally with their surroundings cause simple functions overall patterns to appear” [2]. In cluster of insects, which exist in colonies, such as ants a creature cannot do a work on its own; colony's supportive work is the key motivation shaping the intelligent behavior. The majority of actual ants are blind. Real ants whilst walking arbitrarily leaves a chemical substance known as pheromone [1]. Pheromone magnetizes other ants to stay close to other ants. The pheromone vanishes over time to allow search evaporation. In many experiments presented by Dorigo and Maniezzeo give details about the complex behavior of colonies [3], where ants always prefer shortest path. Figure1. Overview of Dorigo and Maniezzeo experiment [3] Further Parpinelli, Lopes and his associates were the first to propose Ant Colony Optimization (ACO) for learning classification rules, called Ant-Miner. They discover that an ant-based search is more flexible, robust