Flick Pick

Deployed various Machine Learning models to classify the movie reviews as positive or negative.

The movie reviews are present in the documents and then we get the feedback whether the reviews are positive or negative based on the words present in the review.

We're gonna take tokenized words and then use them as features by feeding them into our ML Algorithm. It uses the basics of NLP including Tokenizing, Stemming, Chinking, Chunking and Named Entity Recognition.

We get the probability of the final reviews between 0 and 1 and then we can judge if the review is positive or negative.

Tools Used -

1. Natural Language Processing

2. Sentiment Analysis

3. NLTK

Models Used -

1. tf-idf and Count Vectorizer

2. Support Vector Mechanics

3. Logistic Regression

4. Naive Bayes Classification

Achieved 82.5% accuracy by using the Naive Bayes Classification and using the tf-idf & Count Vectorizer as features.