<![CDATA[Almost every strategic location today has a store that sells daily necessities. These stores compete with each other by offering prices and services that they hope will satisfy their customers. This competition must be anticipated in the management of minimarkets run by cooperatives. Minimarkets run by cooperatives need to maintain the loyalty of their members and general customers by improving service quality. Customer reviews or suggestions and criticism from members or customers are valuable sources of data for evaluating service performance. These customer reviews are unstructured data that are difficult to process manually. This study aims to classify customer opinions on the service quality of cooperative minimarkets into positive, negative, and neutral sentiments using a Lexicon-Based approach. The research methods used are text data preprocessing, sentiment weighting using a lexicon dictionary, classification into positive, negative, or neutral classes, and system performance testing using a confusion matrix. The data labeling stage is carried out automatically using the Lexicon InSet dictionary to determine the sentiment class (positive or negative). The labeled data was then processed using TF-IDF feature extraction and used to train the logistic regression model. Model performance evaluation was carried out using a Confusion Matrix with a training data and test data ratio of 80:20. The results of this study show that the logistic regression algorithm is capable of classifying cooperative service sentiment with an accuracy rate of 81%, precision of 83%, recall of 81%, and an F1 score of 79%. These results indicate that the method used is quite effective in identifying customer opinions and can be used as a decision support system for cooperative managers in continuously improving service quality based on customer sentiment data analysis. ]]>
Copyrights © 2026
