Prediction of Anime Series' Success using Sentiment Analysis and Deep Learning

Viewers' feelings and satisfaction are among the most important influences on movies and other entertainment programs' success. The entertainment industry is one of the most important fields of our time. This paper proposes sentiment analysis to help business leaders make their decisions about investing in this field and the Anime producers to understand viewers' feelings and how satisfied they are. Deep learning techniques are utilized to develop a text sentiment analysis model to classify reviews into positive and negative classes. A data set containing 50,000 reviews from the MyAnimeList dataset is used to train and evaluate a Convolutional Neural Network model. The model is evaluated using different metrics whereby it achieved an accuracy of 0.95, precision of 0.96, recall of 0.98, and F1-score of 0.97.