Abstract

Classification is one of the main topics in data mining or machine learning. Classification is a grouping of data where the data used has a label or target class. Classification is used to collect data and place it into certain groups. The study of the ionosphere is important for research in various domains, particularly in communication systems. In ionosphere research, it is necessary to classify useful and useless radars of the ionosphere. In this paper, we will classify the inosphere data taken from the UCI machine learning repository. Classification is done using three classification methods, namely SVM (Support Vector Machine), Naïve Bayes, and Random Forest. The results of this experiment can show predictions from each experiment with different levels of accuracy and prediction in each method used. The results of the best accuracy, precision, and recall were obtained in the Random Forest method with a ratio of training data and test data of 85%, the accuracy of the test data was 90.57% with a precision of 94.12%.



Keywords - Ionosphere; Classification; SVM; Naïve Bayes; Random Forest.