ENRICHMENT OF ENSEMBLE LEARNING USING K-MODES RANDOM SAMPLING

Abstract
Ensemble of classifiers combines the more than one prediction models of classifiers into single model for classifying the new instances. Unbiased samples could help the ensemble classifiers to build the efficient prediction model. Existing sampling techniques fails to give the unbiased samples. To overcome this problem, the paper introduces a k-modes random sample technique which combines the k-modes cluster algorithm and simple random sampling technique to take the sample from the dataset. In this paper, the impact of random sampling technique in the Ensemble learning algorithm is shown. Random selection was done properly by using k-modes random sampling technique. Hence, sample will reflect the characteristics of entire dataset.

Authors
Balamurugan Mahalingam, S Kannan, Vairaprakash Gurusamy
Madurai Kamaraj University, India

Keywords
Sampling, Ensemble Classifiers, Cluster Random Sample
Published By :
ICTACT
Published In :
ICTACT Journal on Soft Computing
( Volume: 8 , Issue: 1 )
Date of Publication :
October 2017

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.