Classification of Water Pollution with Feature Selection

Abstract

In the feature subset selection problem, a learning algorithm is faced with the problem of selecting a relevant subset of features upon which to focus its attention to achieve the highest predictive accuracy with the learning algorithm on this domain, a feature subset selection method should consider how the algorithm and the training data interact with filter method. This paper applies the normalization by decimal scaling process before feature selection to speed up the learning phase and prevent attributes with initially smaller ranges. This paper uses sequential forward selection to improve the generalization performance of pattern recognizers for water pollute or not. k-Nearest Neighbor classifier is built with filter approach by using the sequential forward selection. To estimate how accurately a classifier labels future data, this paper evaluates the performance of k-Nearest Neighbor classifier on the complete features and the selected feature subset by using the k-fold crossvalidation.