Paper Details

Learning an optimal Bayesian belief network for a Bayesian network classifier is a NP-hard problem. Anumber of heuristic-based algorithms have been proposed for supervised learning of Bayesian belief network such asTree-Augmented Naïve Bayes (TAN). In the past decade, bio-inspired swarm intelligence (SI) based algorithms have beproposed for many optimization problems. Swarm Intelligence based algorithms are characterized by the collectivedecentralized decision-making of several independent agents to search for the optimal solution in the solution searchspace.In this paper, we propose a novel swarm intelligence based Hunting Group Search Bayesian Network ( HuGS) algorithmto learn the Bayesian network for a Bayesian network classifier. HuGS has been inspired by the behavior of a pack ofhunting animals such as wolves. The classification accuracy of the proposed HuGS algorithm is tested against otherBayesian network classifiers such as Naïve Bayes and TAN. The performance of HuGS is evaluated using 20 benchmarkdatasets from the UCI classification datasets. Our experiments show that HuGS outperforms the other state of the artBayesian network learning algorithms.

Shefali K Singhal, “Bio-Inspired Bayesian Network Learning Algorithm for Bayesian Network Classifiers”, International Journal of Advance Engineering and Research Development (IJAERD), Vol. 2, Issue 3, pp. 8-14, March 2015.