Archives

A Novel Clustering based Data Classification Model for Smart Metered Household Electricity Consumption


M. Suresh and Dr.M.S. Anbarasi
Abstract

Global demand of electricity will rise by 85% from 2010 to 2040 and it is due to the increased use of electricity in USA, Europe, China, Japan, Australia and India. Studies reported that the household electricity needs are significantly increasing in the developing countries like India. This paper mainly intends to develop a classification model to analyze the household electricity consumption data using improved ID4 algorithm. But, in recent years, the growth in electricity data urges a requirement to combine the clustering and classification approaches. To overcome the drawbacks of larger size dataset, improved K-means clustering method is incorporated to improved ID3 classifier to attain better classification performance. The presented model is validated on the benchmark dataset from Kaggle and the results are analyzed in terms of with respect to various performance measures. The experimental outcome verified that the clustering process enhances the classification performance and it outperforms the compared methods in a significant way.

Volume 11 | 07-Special Issue

Pages: 286-295