Retracted: Classification of High Dimensional Data Using Filtration Attribute Evaluation Feature Selection Method of Data mining

Feature selection is the process to reduce the dimension of the high dimensional data which reduces the time complexity of any classifier and to increase the predictive precisions from the classifier and also consider the effect from the suitable features on here preprocessing of the data is the crucial stage to form the correct and quality dimension for the classification. Therefore feature selection can be classified into two types one is forward selection and another is backward selection many research applications select any one of the method to classify the data here we are using the both backward and forward selection to improve the performance of the classifier after applying the feature selection methods. So in this work we proposed different feature selectors followed by different classification algorithms are applied and also with accuracy of the classifier in this the standard bench dataset UCI and Ken ridge biomedical data repository are used to estimate the classifier accuracy of different datasets. We applied the different classification algorithms like J48, Naïve bayes, k-star, Random forest for calculating the accuracy their comparison among with the feature selection averages.