An Improved Flexible Partial Histogram Bayes Learning Algorithm

This paper presents a proposed supervised classification technique namely flexible partial histogram Bayes (fPHBayes) learning algorithm. The traditional classification algorithms like neural network, support vector machine, first nearest neighbor, nearest subclass classifier and Gaussian mixture model classifier are accurate but slow when dealing with large number of instances. In additional to that these algorithms might require to be retrain when the classes changes. On the other hand, algorithms like naïve Bayes and nearest class mean are fast but not accurate. It is difficult and challenging to have a classification algorithm that is fast and accurate when dealing with large number of instances. In our previous work, partial histogram Bayes (PHBayes) learning algorithm showed some advantages in the aspects of speed and accuracy in classification tasks. However, its accuracy declines when dealing with small number of instances or when the class feature distributes in wide area. In this work, the proposed fPHBayes solves these limitations. fPHBayes is able to work fast with good accuracy with large and small number of instances. fPHBayes uses a probability distribution derived from smoothing the observed histogram in order to represent the class. Then it performs the classification using the Bayesian rule. fPHBayes was analyzed and compared with PHBayes and other standard learning algorithms like first nearest neighbor, nearest subclass mean, nearest class mean, naive Bayes and Gaussian mixture model classifier. The experiments were performed using both real data and synthetic data considering different number of instances and different variances of Gaussians. The results showed that fPHBayes is more accurate and flexible to deal with different number of instances and different variances of Gaussians as compared to other classifiers.