<![CDATA[Dimension reduction is a method applied in machine learning sector to significantly improve the efficiency of computational process. The application of high number variables in certain dataset is expected to be able to provide more information to analyze. However, this application of high number of variables will impacted on the computational time and weight linearly. Dimension reduction method serves to transforming the high dimension data into much lower dimension without significantly reduce the initial information and characteristic provided by the initial data. Core and Reduct is a method acquired through the concept of Rough Set. Dataset functioning as the input and output on Machine Learning can be perceived as informational system. The objective of this research is to determine the impact of the dimension reduction application on machine learning algorithm on the reduction of computational time and weight. Core and Reduct will be applied in few popular machine learning method such as Support Vector Machine (SVM), Logistic Regression, and K-Nearest Neighbors (KNN). This research applied on 5 UCI machine learning dataset which are Iris, Seeds, Years, Sonar, and Hill-Valley. Furthermore, Machine learning metrics such as Accuracy, Recall, Precision, and F1-Score also observed and compared. This research resulted in the conclusion that Core and Reduct is able to decrease the computational time up to 80% and maintain the value of each evaluation model.]]>
