<![CDATA[The existence of the MRT Jakarta is expected to reduce the number of private transportation uses, which causes the congestion rate in the Jakarta area to continue to increase. Reviews from MRT Jakarta users help MRT Jakarta in improving its services, because good service quality can attract people to use MRT as public transportation for traveling. However, at this time, MRT Jakarta's official social media accounts had not yet found a feature to sort reviews between positive and negative reviews. If this is done manually it will take time, therefore it is necessary to carry out an automation process in the selection of these reviews. This automation process is known as sentiment analysis. In this study, the sentiment analysis system uses a combination of the Neighbor-Weighted K-Nearest Neighbor (NWKNN) classification method with the Information Gain feature selection. Tests conducted in this study using 5-Fold Cross Validation. The test results reach the optimal point at the 5th Fold, when the k value = 100, the exponent value = 2, and the threshold value for feature selection = 100% (without feature selection and without using stopword removal), with values of precision, recall, f- measure, and accuracy is 1; 0.94; 0.97; and 0.97.]]>
Copyrights © 2020
