K nearest neighbor reinforced expectation maximization method


ACI M., AVCI M.

EXPERT SYSTEMS WITH APPLICATIONS, vol.38, no.10, pp.12585-12591, 2011 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 38 Issue: 10
  • Publication Date: 2011
  • Doi Number: 10.1016/j.eswa.2011.04.046
  • Journal Name: EXPERT SYSTEMS WITH APPLICATIONS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.12585-12591
  • Keywords: K nearest neighbor method, Bayesian method, Expectation maximization algorithm, Hybrid method, Classification, Clustering
  • Çukurova University Affiliated: Yes

Abstract

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. (C) 2011 Elsevier Ltd. All rights reserved.