Predictive Modelling of Movie Performance Using Supervised Learning and IMDb-Based Feature Analysis

Abstract

The movie industry requires substantial financial resources, making it crucial to anticipate a film’s commercial success. This study investigates a machine learning-based method to categorize movies as HIT, AVERAGE, or FLOP using their IMDB ratings. The analysis employs three machine learning algorithms—Naïve Bayes, Logistic Regression, and Support Vector Machine (SVM)—on a realworld dataset obtained from Kaggle, which includes 5043 records with 23 attributes. The dataset is preprocessed by handling missing values, encoding categorical variables, and splitting it into training and testing sets. Experimental results indicate that Logistic Regression achieves the highest accuracy of 99%, followed by SVM with 79%, and Naïve Bayes with 35%. Furthermore, a user-friendly graphical interface has been developed, enabling users to upload datasets, train models, predict the success of new movies, and compare algorithm performance. The findings highlight Logistic Regression as a highly reliable approach for forecasting movie performance, providing a valuable decision-support tool for producers and investors.