Volume 2 - Issue 3, May - June 2026

The film industry is one of the most uncertain and competitive sectors where substantial investments do not always guarantee commercial success. Movie performance depends on multiple interconnected factors such as production budget, marketing expenditure, audience engagement, actor popularity, director reputation, ticket pricing, and release scale. Traditional methods of evaluating movie success are often based on intuition, experience, and historical observations, which may not accurately capture the complex relationships among these variables. This research presents a Machine Learning-based Movie Success Prediction System designed to classify movies as either HIT or FLOP while simultaneously evaluating their financial performance. Due to the limitations of publicly available datasets, a structured synthetic dataset was developed using logical relationships between critical movie-related features. The proposed system employs a Random Forest Classifier to learn patterns from historical and simulated data and generate prediction outcomes. In addition to classification, the system performs financial analysis by estimating revenue, total cost, and profitability based on audience occupancy, seating capacity, ticket price, and marketing investment. A Streamlit-based dashboard was developed to provide an interactive and user friendly platform where users can experiment with different movie scenarios and observe both predictive and financial outcomes in real time. Experimental evaluation demonstrated an overall accuracy of 86.40%, with a precision of 90%, recall of 74%, and F1-score of 81%. The results indicate that integrating machine learning prediction with financial evaluation provides a more comprehensive decision-support framework than traditional prediction-only systems. The proposed approach can assist producers, investors, and distributors in making more informed decisions during movie planning and release stages.

Shivam Yadav, Sandhya Kaprawan,"Movie Success Prediction Using Machine Learning and Financial Performance Analysis" International Journal of Advanced Multidisciplinary Research and Educational Development, V2(3): Page(914-921) May-June 2026. ISSN: 3107-6513. www.ijamred.com. Published by Scientific and Academic Research Publishing.