Feature Extraction and Classification of Movie Reviews using Advanced Machine Learning Models

The usage of the internet has grown in different systems since the mid-90s. Individuals express various comments on the internet depending on their emotions. Twitter is the most well-established small-scale blogging. It has 330 million diverse users every month and 145 million daily active users. For this research, we're attempting to evaluate the thoughts of the users in the IMDB movie reviews on tweets obtained from different outlets. The experiment forecasts correctly that the analysis feeling would be of a negative or optimistic polarity. We came across various approaches like word2vec, Doc2vec for matching and SVM, ANN, RF for classification, and defined different strategies from which we could identify the features of the reviews either considering it as a single entity or considering it as a whole text. We also applied a methodology named distributed representation of sentences and records, which eventually gives us the highest precision of all the models introduced.