<![CDATA[Customer satisfaction serves as a crucial indicator for assessing the quality of after sales services. PT AUTO2000 Sukun routinely gathers customer feedback through satisfaction surveys. However, the manual evaluation process is both time consuming and susceptible to subjective judgment. This study seeks to classify customer satisfaction based on customer comments using the Multinomial Naive Bayes algorithm. The dataset comprises 108 customer reviews accompanied by manually assigned positive and negative sentiment labels. The research procedure consisted of text preprocessing, including case folding, tokenization, stopword removal, and stemming, followed by feature extraction using Term Frequency Inverse Document Frequency (TF IDF). The dataset was then divided into 70% training data and 30% testing data. Synthetic Minority Over sampling Technique (SMOTE) was applied exclusively to the training data to alleviate class imbalance. Model performance was assessed using accuracy, precision, recall, F1 score, and a confusion matrix. The findings reveal that the Multinomial Naive Bayes algorithm attained an accuracy of 76 percent. For the negative class, the model achieved a precision of 79%, a recall of 86%, and an F1 score of 83%. For the positive class, it obtained a precision of 67%, a recall of 55%, and an F1 score of 60%. These findings suggest that the Multinomial Naive Bayes algorithm exhibits satisfactory capability in classifying customer satisfaction. Nevertheless, the model still encounters challenges in accurately recognizing all instances belonging to the positive class.]]>
Copyrights © 2026
