Predictive Modeling of Animation Movie Dataset: Revenue Prediction and Feature Analysis

The animation film industry has grown significantly over the past few decades, increasingly generating substantial global revenue; however, predicting a movie’s revenue remains challenging due to numerous influencing factors, such as budget, genre, production company, and language. This study analyzes a comprehensive animation movie dataset to identify key factors contributing to revenue generation. The primary objective is to develop a machine learning model that can predict revenue, enabling producers and stakeholders to make data-driven decisions in planning and investment. Our analysis involves extensive exploratory data analysis to uncover trends and correlations, such as the positive relationship between budget and revenue and the high popularity of genres like adventure and family. After preprocessing the data to handle missing values and encoding categorical variables, we employed a random forest regressor to predict movie revenue. The model achieved a mean squared error of 506,733,789,216,286.4, indicating the difficulty of accurately forecasting movie revenue due to the high data variance. We also implemented a binary classification of revenue levels (high and low) and evaluated the model’s performance using a confusion matrix and receiver operating characteristic curve. We determined an area under the curve score of 0.87, reflecting moderate predictive power. The results highlight that budget, genre, and production company significantly affect revenue. These findings provide valuable insights for industry stakeholders, as understanding these revenue-driving factors can improve decision-making in developing, marketing, and distributing animation films. Further research could explore advanced models, feature engineering, and additional data sources to enhance prediction accuracy and broaden the understanding of revenue determinants in the animation industry.