Bosca Living, a star seller on Shopee and Tokopedia, is facing the challenge of customer sentiment analysis. This research evaluates models and methods to strengthen the response to customer feedback. In previous studies, feature extraction techniques such as Term Frequency-Inverse Document Frequency (TF-IDF), Word2Vec, FastText, and Global Vectors for Word Representation (GloVe) have been tested. Machine learning models such as K-Nearest Neighbors (KNN), Random Forest, Support Vector Classifier (SVC), XGBoost, Logistic Regression, and Decision Tree have been employed, but a more in-depth comparison is needed according to Bosca Living's assessment. This research proposes a model comparison through preprocessing, feature extraction, and parameter determination stages using GridSearchCV. Machine learning models like KNN, Random Forest, SVC, XGBoost, Logistic Regression, and Decision Tree are evaluated with StratifiedKFold to reduce the risk of overfitting. The research results provide deep insights, guiding Bosca Living in improving responses to customer feedback. This approach is expected to optimize business strategies, support continuous improvement, and be responsive to market dynamics and evolving customer needs