K nearest neighbor reinforced expectation maximization method

K nearest neighbor and Bayesian methods are effective methods of machine learning. Expectation maximization is an effective Bayesian classifier. In this work a data elimination approach is proposed to improve data clustering. The proposed method is based on hybridization of k nearest neighbor and expectation maximization algorithms. The k nearest neighbor algorithm is considered as the preprocessor for expectation maximization algorithm to reduce the amount of training data making it difficult to learn. The suggested method is tested on well-known machine learning data sets iris, wine, breast cancer, glass and yeast. Simulations are done in MATLAB environment and performance results are concluded.

► A data elimination approach is proposed to improve data clustering.
► Based on hybridization of k nearest neighbor and expectation maximization algorithms.
► Data that make difficult to learn are eliminated to achieve successful results.
► Proposed method shows better classifying performance than compared classifiers.
► The hybrid method is applicable to noisy data set classifying applications.