Early-production stage prediction of movies success using K-fold hybrid deep ensemble learning model

The Indian movie industry is the largest movie industry based on the number of movies produced per year. It is also the most diverse movie industry.It has been examined in a recent study that only a few of the movies achieved success. Revenue uncertainties have created immense pressure on the motion picture industry. Researchers and film producers continually feel a necessity to have some expert systems that predict the movie's success probability preceding its production with reasonable accuracy. The diversity of the Indian movie industry makes the problem more difficult. Only a few researchers worked on Indian films, but most of them are targeted prerelease forecasting or have low prediction accuracy. This study focused on Indian movies and concentrated on the upcoming film's success as soon as a quotient (director, cast) signed an agreement. This proposed forecasting has been considered as the earliest forecasting. Our study retrieved and used the last 30 years of Indian movie information covering all India's regional movies.We had judicially chosen some of the movie's intrinsic features and introduced a set of novel derived features to increase the forecasting accuracy. We had proposed a K-fold Hybrid Deep Ensemble learning Model (KHDEM), which includes Deep Learning models (DLM) and ensemble learning models. Finally, We made the prediction using a Logistic Regression (LR) classifier. We had implemented a binary classification model and achieved 96% accuracy, which outperforms all the benchmark models. The introduction of our derived features had improved the accuracy by 17.62%.This study highlights the potential of predictive and prescriptive data analytics in information systems to support industry decisions.

Automated summary

The Indian movie industry is the largest and most diverse industry in terms of the number of movies produced per year. Only a few movies achieve success, and revenue uncertainties create immense pressure on the industry. This study focuses on predicting the success of upcoming Indian movies through a K-fold Hybrid Deep Ensemble learning Model (KHDEM) that combines Deep Learning models and ensemble learning models. With the implementation of a Logistic Regression classifier, the study achieved 96% accuracy, outperforming all benchmark models. The introduction of derived features improved the accuracy by 17.62%. This study highlights the potential of predictive and prescriptive data analytics in supporting industry decisions.