<![CDATA[General Background: The rapid growth of e-commerce platforms has intensified the need to understand consumer sentiment to improve service quality and competitiveness. Specific Background: ZALORA, as a leading online fashion retailer in Southeast Asia, has accumulated vast user-generated feedback, particularly on platforms like the Google Play Store. Knowledge Gap: Despite the availability of such data, limited studies have analyzed consumer sentiment using machine learning methods specifically tailored to ZALORA’s mobile platform. Aims: This study aims to examine consumer sentiment toward ZALORA products and assess the effectiveness of the Random Forest algorithm in classifying sentiment. Results: Utilizing a quantitative approach, 1,200 user reviews were analyzed, with 63.5% expressing positive sentiment. Word cloud visualization supported this finding, revealing frequently mentioned terms such as “product,” “goods,” and “shopping.” The Random Forest model achieved an accuracy of 80%, with precision, recall, and F1-score values for positive sentiment all exceeding 0.80. Novelty: This research integrates TF-IDF-based preprocessing with Random Forest classification to enhance sentiment analysis performance specifically for mobile commerce reviews. Implications: The findings highlight the potential of machine learning in extracting actionable insights from user reviews, offering practical implications for improving customer experience and guiding strategic development in digital retail platforms. Highlights: High accuracy (80%) achieved using Random Forest for sentiment classification. TF-IDF preprocessing significantly improved model performance. Word cloud analysis revealed key satisfaction indicators from users. Keywords: Sentiment Analysis, Random Forest Method, Zalora Consumers ]]>
