In recent years, data have become increasingly larger in both number of instances and number of features in many applications. This enormity may cause serious problems to many machine learning algorithms with respect to scalability and learning performance. Therefore, feature selection is essential for the machine learning algorithms while handling high dimensional datasets. Many traditional search methods have shown promising results in a number of feature selection problems. However, as the number of features increases extremely, most of these existing methods face the problem of intractable computational time. Since no single feature selection method could handle all requirements of feature selection in real world datasets, hybrid methods prsented here are the tested methods for effecive Feature Selection.One viable option is to apply a ranking feature selection method to obtain a manageable number of top ranked features which could be further handled by traditional feature selection methods for further analysis.