Abstract

This study presents a model for collecting and analyzing customer reviews using machine learning. As e-commerce platforms grow, large volumes of customer opinions are generated, but most of this data is unstructured and complex to use directly. To address this, the proposed approach begins with gathering reviews from online platforms, followed by data preprocessing to clean, filter, and normalize the text. These steps ensure that the data is suitable for machine learning analysis. Classification models are then applied to perform sentiment analysis. The results show that the model can categorize customer opinions with an accuracy of over 94%. The findings are illustrated through comprehensive graphs and charts, offering clear insights into customer attitudes and behavioral patterns. Such visualizations are crucial in translating complex analytical results into an accessible form, enabling managers to identify key trends and underlying issues rapidly. By presenting data from multiple perspectives, the visual outputs facilitate more profound understanding, support comparative analysis, and enhance the overall quality of decision-making, thereby strengthening evidence-based management practices within dynamic business environments. The study highlights that online customer feedback reflects individual opinions and influences overall business performance. By systematically analyzing customer feedback, companies can identify emerging needs, enhance products and services, and design more effective business strategies. Such insights directly strengthen competitiveness in the e-commerce sector, where user experience is a critical differentiator. Beyond e-commerce, the proposed methodology demonstrates flexibility and applicability across other domains, including finance, healthcare, and education, where capturing and interpreting customer sentiment is equally important. This research establishes a robust and practical framework for transforming unstructured feedback into actionable knowledge by integrating machine learning techniques with multi-dimensional analytics. The framework supports strategic decision-making and provides organizations with a sustainable approach to leveraging data for innovation and long-term value creation.