Sentiment Analysis of IMDb Movie Reviews : A comparative study on Performance of Hyperparameter-tuned Classification Algorithms

Sentiment Analysis (SA) is a sub-domain of Natural Language Processing where useful insights on sentiment and opinion of people can be obtained and analyzed from various textual data in structured, unstructured and semi-structured format. In this work, I have tried to analyze the sentiment of viewers from the IMDb Movie Reviews dataset. For this, I have taken three different Supervised learning methods, namely, Linear Support Vector Machine, Logistic Regression and Multinomial Naive Bayes Classifier, each with different settings of hyperparameters. Moreover, to capture the notion of informal jargon, approach of N-grams has been followed. Furthermore, a comparative study has been performed to find the one best model from each of the different types of above mentioned Supervised learning techniques based on their Accuracy Score, F1-Score and AUC Score. In this approach, I have obtained the best accuracy score of around 0.910 and mean F1-score after 10-fold Cross validation of around 0.894.