Analisis Sentimen Pada Rating Aplikasi Shopee Menggunakan Metode Decision Tree Berbasis SMOTE

Abstrak

Text mining research that performs the categorization procedure on text documents is known as sentiment analysis. Sentiment analysis is the process of extracting a person's ideas, emotions, and evaluations expressed about a certain issue using natural language approaches. Researchers used the Decision Tree approach to do sentiment analysis on the Shopee application rating. The goal of this research is to find out how accurate this Shopee app is and what users think about it. The accuracy value of 99.91 percent, AUC (Area Under Curve) 0.999, recall 99.88 percent, and precision value 99.98 percent were obtained using the Decision Tree method using SMOTE (Synthetic Minority Oversampling Technique). The accuracy of the Decision Tree method without SMOTE is 99.89 percent, the AUC (Area Under Curve) is 0.950, the recall is 99.88 percent, and the precision is 99.98 percent. Based on the findings of the current study, it can be concluded that SMOTE has an effect on value accuracy and AUC (Area Under Curve), but has no effect on recall and precision values, and that the results are the same whether SMOTE is used or not. The accuracy value achieved differs by 0.02 percent, whereas the AUC differs by 0.049