Description

Notes:
● This assignment has one part – Naive Bayes.
● Submit whole code with a detailed write-up.
● The report should be a brief description explaining what you did. Include any observations and/or plots required by the question in the report. ● You should use Python for all your programming solutions.
● Your assignments will be auto-graded, make sure you test your programs before submitting. We will use your code to train the model on training data and predict on test set.

1. Text Classification using Naive Bayes
In this problem, we will use the Naive Bayes algorithm for text classification. The dataset for this problem is a subset of IMDB movie review data. First entry of each row is a review followed by a sentiment. The sentiment attached with the review has two class labels: positive and negative. Given a review, task is to predict the sentiment of the review. Take positive label as 1 as negative label as 0
(a) (50 points) Implement the Naive Bayes algorithm to classify each of the articles into one of the given categories.The output should contain the predicted label of the review (positive or negative) in terms of 1 and 0 in a text file format. Report the accuracy over the training as well as the test set. Notes: I. Use Unigrams as features.
II. Make sure to use the Laplace smoothing in Naive Bayes to avoid any zero probabilities. Use c=1.
III. You should implement your algorithm using logarithms to avoid underflow issues.
IV. You should implement Naive Bayes from the first principles and not use any existing python modules.

Perform stopword removal and stemming before using naive bayes. You can use libraries ( nltk for python) to perform stopword removal and stemming.