查看文章

Data mining is playing a significant role in the digital era, and there are traditional techniques to classify, cluster the large data, etc. Today, the variety of data and its size has grown increasingly. Preprocessing of the data impose and need high computational resources due to raising the number of data attributes. Thus, attributes reduction deem a vital and significant part of the data pre-processing due to its ability to reduce the required computational resources. In this study, a hybrid model is proposed to eliminate irrelevant attributes with N number of goodness evaluation metrics by using K-Means Clustering and Bagging Ensemble Classifier. The proposed model was implanted with five different datasets. The model can minimize the number of the attributes up to (70%). Hence, the results with reduction can be increased the efficiency of the classification performance from the computation time standpoint.