Sentiment Analysis Using Machine Learning Techniques on IMDB Dataset

Abstract

Artificial intelligence is utilized in various sectors due to the surge in data. The fundamental element of artificial intelligence is data, which is sourced from various channels, including websites. It involves analyzing emotions, opinions, and attitudes expressed in text data. Websites provide significant volumes of comment data, which is used to conduct sentiment analysis studies. Sentiment analysis is a natural language processing (NLP) task, which is a core focus of artificial intelligence research. Sentiment analysis was conducted on open-source data obtained from IMDb, a platform containing information about films as well as reviews of actors, directors, and movies. The data was preprocessed before word vectorization was performed with the frequency-based TF-IDF method, and then machine learning algorithms were employed for classification. SVM was found to have the highest accuracy of 90%. © 2023 IEEE.