COMPARISON OF INFORMATION GAIN AND GAIN RATIO IN K-NEAREST NEIGHBOR METHOD

Abstract

Feature selection is one of the methods used to reduce or reduce the dimensions between attributes so as to produce relevant attributes, this method is considered to be able to reduce anomalies in attributes in a data set. The KNN method is considered a classification method which is classified as a method with low accuracy, especially when compared to other classification methods. Farid Naufal's research, 2021 the accuracy of the KNN method gets a value of 75.4% while the SVM method gets a value of 85.7%. The solution to the problem of the small value of the classification is using feature selection, this technique is used to reduce the dimensions of the attribute. This research uses Information Gain feature selection technique and its development, namely Gain Ratio. The gain ratio technique is considered better when compared to the weighting on the glass data, it can be seen that the Gain Ratio weighting value is better than its predecessor Information Gain.