The program is implemented using the 'Naive Bayes Algorithm' and is trained and tested using a subset of the Pang and Lee 2002 Sentiment Classification / Movie Review data, created by Ted Pedersen (tpederse@d.umn.edu). This program yields 80.8% accuracy based on those testing and training data.
Speech and Language Processing An Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition, Third Edition draft (1/2022), Daniel Jurafsky and James H. Martin.