A Comparative Study on Feature Selection for E

This paper explores the application of feature selection by the Correlation based Feature Selection (CFS) algorithm on the problem of classification of E.coli promoters using neural networks, Support Vector Machines (SVM) and Extreme Learning Machines (ELM). It was found that even though in general the classification accuracy can be reduced by a statistically significant amount, in real terms this was only a few percent. The results also indicate some interesting characteristics of the features used in E-coli promoters. A comparative study with three typical classifiers was carried out in this study. Keyword: E.coli, promoters, classification, pattern recognition, neural networks, SVM, ELM, feature selection