Machine Learning-Based Classification for Sentimental Analysis of IMDb Reviews

Filmmakers and audiences both need to understand how the public feels about movies in the age of digital media, which is ruled by sites like IMDb. Given the volume of information available on IMDb, research into how machine learning algorithms can consistently classify movie reviews as positive or negative attitudes is required. This study aims to measure the way various machine learning algorithms perform when sentiment analysis of IMDb movie reviews is performed to gain knowledge about the preferences and responses of the audience. An IMDb movie review dataset was subjected to a range of machine learning techniques, such as k-nearest neighbors (KNN), Naive Bayes, logistic regression, random forest, and decision trees. The dataset was used to train and test each algorithm's ability to categorize views as positive or negative. Among the methods under study, logistic regression achieved the highest accuracy at 88.5%, closely followed by Naive Bayes at 85%. Compared to random forests, which achieved an accuracy of 85.2%, decision trees performed relatively well, with an accuracy of 74.5%. However, KNN only demonstrated mediocre effectiveness, with an accuracy of roughly 75.5%. These results show how effectively sentiment analysis of IMDb movie reviews using logistic regression performs. The findings demonstrate how important sentiment analysis is to understanding user opinions about movies on websites such as IMDb. Filmmakers and industry observers can employ machine learning techniques, particularly logistic regression, to gain insightful information about the emotions and preferences of their audience.