<![CDATA[The entertainment industry faces high financial risks in movie production, while large-scale online data such as IMDb provide opportunities for predictive analysis. This study aims to apply machine learning, specifically the Naïve Bayes algorithm, to predict movie success and analyze emerging audience trends through a Decision Support System (DSS). Recent IMDb datasets (2022–2025) were collected and preprocessed. Movies were classified into “Successful” and “Less-successful” categories based on audience ratings and vote thresholds. The Naïve Bayes algorithm was chosen for its efficiency and interpretability in handling textual and categorical data. Model performance was evaluated using accuracy, precision, and recall. The model achieved 95.7% accuracy with balanced precision and recall. Trend analysis showed that Action movies consistently had the highest likelihood of success, while genres such as Crime and Biography demonstrated moderate performance. The DSS framework provided useful insights for producers and distributors in reducing risks. Naïve Bayes proved effective for sentiment-driven prediction when embedded in a DSS, supporting strategic decision-making in the film industry. Despite limitations in feature independence and reliance on IMDb, the study shows the potential of machine learning–based tools to align production with audience preferences. Future research should expand to multi-platform datasets and advanced models to enhance predictive robustness and industry applicability.]]>
