Amazon Review Classification using NLP

Table of Contents

• About the Project
  • Steps
  • Built With
• Fork the Repo and Contribute
• Contact

About the Project

In this Classification Project, we use Natural Language Processing techniques and different Machine Learning models to classify Amazon book reviews into POSITIVE or NEGATIVE sentiments.

Steps

1. Bag of Words technique was used to vectorized the reviews using the CountVectorizer (binary only) and TfidfVectorizer (Term Frequency - Inverse Document Frequency).

2. Since the smaller dataset was imbalanced, we use a bigger dataset and balance out the POSITIVEs and NEGATIVEs.

3. The models that were compared are as follows:

   • Linear SVM
   • Decision Tree
   • Naives Bayes
   • Logistic Regression

4. Further, GridSearchCV was used to find the best parameters for the Linear SVM model.

5. The Linear SVM model with best parameters found using GridSearchCV is stored using the Pickle library for future use.

6. In future, other models will be studied for finding best parameters of for each. An overall best performing model can then be found.

Built With

• Jupyter Notebook
• Scikit-learn
• Pickle
• Pandas
• json

Fork the Repo and Contribute

Contributions are what make the open source community such an amazing place to be learn, inspire, and create. Any contributions you make are greatly appreciated.

1. Fork the Project (click on Fork in the top-left corner)
2. Create your Feature Branch (git checkout -b feature)
3. Commit your Changes (git commit -m 'Add some AmazingFeature')
4. Push to the Branch (git push origin feature)
5. Open a Pull Request

Contact

Sinjoy Saha

• LinkedIn
• Twitter