Naive Bayes Classifier

Naive Bayes Classifier

Github

Introduction

Naive Bayes methods are a set of supervised learning algorithms based on applying Bayes’ theorem with the “naive” assumption of conditional independence between every pair of features given the value of the class variable.

• 1. Imports.
• • 
• 2. Reading Datasets Amazon, Yelp, Kaggle.
• 
• 
• 
• 
• 3. Cleaning the dataset.
• 
• 
• 4. Positive sentiments vs Negative sentiments graph.
• 
• 
• 5. Split data into train, test and dev.
• 
• 
• 
• 
• 
• 6. Build Vocabulary
• 
• 7. Probability of the occurrence of the words.


• 
• 8. Calculate accuracy using dev dataset.
• 
• 9. Conduct five fold cross validation.
• 
• 
• 
• 10. Compare the effect of Smoothing.
• 
• 
• 
• 11. Derive Top 10 words that predicts positive and negative class.
• 
• 12. Final Accuracy.
• 
• I could achieve an accuracy of 73.81%

• Graphs
• We can see from the above graphs that the positive and negative reviews are almost the same.

• Challenges
• I faced a challenges in computing the probability values and accuracies which I could overcome with the help of a few references.

• References
• https://www.karanr.dev/blog/2020/naive_bayes_large_movie_reviews/naive_bayes_large_movie_reviews/
• https://cs230.stanford.edu/blog/split/
• https://levelup.gitconnected.com/movie-review-sentiment-analysis-with-naive-bayes-machine-learning-from-scratch-part-v-7bb869391bab
• https://satyam-kumar.medium.com/imdb-movie-review-polarity-using-naive-bayes-classifier-9f92c13efa2d
• https://scikit-learn.org/stable/modules/naive_bayes.html