Multi-Relational Stacking Ensemble Recommender System Using Cinematic Experience

With the advent of many movie content platforms, users face a flood of contents and consequent difficulties in selecting appropriate movie titles. Although much research has been done in developing effective recommender systems to provide personalized recommendations based on customers' past preferences and behaviors, not much attention has been paid to leveraging users' sentiments and emotions together. In this study, we built graph-based movie recommender system that utilized sentiment and emotion information along with user ratings, and evaluated its performance in comparison to well known conventional models. The sentiment and emotion information was extracted using fine-tuned BERT. We used a Kaggle dataset created by crawling movies' meta-data and review data from the Rotten Tomatoes website. The study results show that the proposed models coupled with emotion and sentiment are superior over the conventional models. The results of this study support the possibility of using sentiment and emotion information together in relation to movie recommender system.