A Novel Approach for Feature Selection

Clustering problem suffers from the curse of dimensionality. Dimensionality reduction of a feature set refers to the problem of selecting relevant features which produce the most predictive outcome and similarity functions that use all input features with equal relevance may not be effective. We introduce an algorithm that discovers clusters by different combinations of dimensions via local weightings of features. This approach avoids the risk of loss of information encountered in global dimensionality reduction techniques, and does not assume any data distribution model. Our method associates to each cluster a weight vector, whose values capture the relevance of features within the corresponding cluster. To judge the efficiency of the proposed method the results are experimentally compared with other optimization methods such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO) and Differential Evolution (DE) for feature selection.

Authors and Affiliations

1. Computer Science Department, JNTU Kakinada, Kakinada, Andhra Pradesh, India

   Ch. Swetha Swapna, V. Vijaya Kumar & J. V. R. Murthy

Corresponding author

Correspondence to Ch. Swetha Swapna .

Editors and Affiliations

1. University of Kalyani, Kalyani, West Bengal, India

   J. K. Mandal

2. Department of Computer Science and Engineering, Anil Neerukonda Institute of Technology and Sciences, Vishakapatnam, India

   Suresh Chandra Satapathy

3. Dean, Faculty of Engineering, Technology, University of Kalyani, Kalyani, West Bengal, India

   Manas Kumar Sanyal

4. Engineering and Technological Studies, University of Kalyani, Kalyani, West Bengal, India

   Partha Pratim Sarkar

5. Department Computer Science & Engineering, University of Kalyani, Kalyani, India

   Anirban Mukhopadhyay

Reprints and permissions

© 2015 Springer India

Policies and ethics