Feature selection is an important term in machine learning tasks as it can efficiently improve the performance of the model by eliminating the redundant and irrelevant attributes. Feature selection not only improves the quality of the model, it also makes the process of modeling more efficient. Due to irrelevant and huge dimensions data the quality of the model may degrade. This paper presents the importance of feature selection on various classification algorithms. In this study, the feature

more »

... lection techniques like attribute evaluator and the best first search is used for reducing the number of features. The dimension is reduced from eight to four. The dataset used is Pima diabetic dataset from UCI repository. The substantial increase is given in terms of accuracy.