Increasing the efficiency of Naïve Bayes model for Sentiment Analysis
- Factored in feature extraction techniques, Negation Handling and Word N Grams, trained model on IMDB data
- Obtained accuracy of 85.1% for linear complexity model, at par, to the support vectors machines and neural models