Decision functions estimation using Inclined Planes system Optimization algorithm

The classification problem is posed as one of the most important issues in pattern recognition and its great applications in sciences and engineering has attended researches. In this paper, a classifier based on new algorithm of Inclined Planes system Optimization (IPO) is proposed. The proposed classifier estimates the decision hyperplanes for separating the feature space using algorithm of inclined planes system optimization. This algorithm has been used in several optimization problems and its ability to find optimal solution has been proven. However, in recent studies, the inclined planes system optimization algorithm has not been used to estimate the decision functions in feature space to classify the data. The performance of the proposed classifier is evaluated by the data sets of Iris, Wine, Breast Cancer and Liver Disorders from UCI machine learning repository. The comparative results show more potency of the proposed classifier than other classifiers based on the metaheuristic such as particle swarm optimization (PSO) and genetic algorithm (GA).