Sentiment analysis helps the hotel industry understand customer preferences and experiences by categorizing reviews as positive, negative, or neutral. In Kuala Lumpur, where the hotel sector remains highly profitable, online reviews significantly influence business strategies. Platforms like TripAdvisor play a crucial role in shaping customer perceptions and decision-making. This study seeks to develop a sentiment analysis model using machine learning techniques, complemented by an interactive visualization dashboard, to analyse customer reviews of hotels in Kuala Lumpur. The primary goal is to classify customer reviews to gain insights into customer preferences. The Kaggle dataset, based on TripAdvisor reviews is collected and pre-processed. Feature extraction approaches, including Word2Vec embeddings and TF-IDF are applied. The classification of hotel reviews is then performed using two machine learning techniques, which are the Decision Tree and Naive Bayes classifiers. The data is split into training and testing sets with varying ratios such as 60:40, 70:30, and 80:20. The model’s performance is evaluated using key metrics such as accuracy, precision, recall, and F1-score. The best model performance is attained using the TF-IDF feature extraction method in combination with the Naïve Bayes classifier, achieving an accuracy of 95.14%. This study helps the hotel industry to understand how online reviews affect hotel reputations and support strategic decision-making. Future research should aim to enhance the model by incorporating multilingual sentiment analysis, enabling real-time sentiment detection, and utilizing ensemble methods that integrate multiple machine learning and deep learning models to improve accuracy.