SENTIMEN ANALISIS APLIKASI MOTORKU X MENGGUNAKAN METODE NAIVE BAYES CLASSIFIER

The rapid development of technology can bring convenience to humans in living life. Technology that continues to develop certainly produces very large amounts of data. The data can provide useful information if it can be processed properly. The Motorku X application is one of the innovations created by Astra Motor to make it easier for consumers or potential customers to service and purchase motorbikes. The Motorku X app generates review data on a daily basis. The review data can be used as application development for the future. In order for reviews to be put to good use, review data needs to be processed and analyzed first. The method for conducting data analysis is sentiment analysis. The classification method used in this study is the naive Bayes classifier method. The stages carried out in this study were data collection, labeling, pre-processing, split data, tf-idf weighting, classification, and evaluation. The amount of data used in this study was 1000 data and divided into two classes, namely positive and negative classes. The research was conducted with 3 training and testing data sharing scenarios, namely 90%:10%, 80%:20%, and 70%:30%, producing the best results at a ratio of 90%:10% with 76% accuracy, 76% precision , and recall 97%. Keywords: Analysis Sentiment, Naive Bayes, Motorku X.