cmdenys / goodreads

Text analysis of book descriptions from Goodreads

A "Goodread" or a Bad Read?

A machine learning project to predict book ratings based on their descriptions.

📈 Training the Model

We trained our model using data from Goodreads scraped by Naren on Kaggle.com. The data is skewed to higher performing books on Goodreads.

NOTE: Goodreads discontinued its API functionality so this model is frozen in 2019's best books of Goodreads.

📚 The Model in Action

Check out our app to see the model in action. Paste the description of your favorite books in the search bar to see if it's rated good or bad. (Give genres a try too, just for kicks.)

💾 Installation

Please refer to the requirements.txt file to be sure you have the necessary tools and libaries to run our code locally.

Once you have everything installed, run the following in your terminal: python app.py

💻 Contributors

• Charlie Denys - Machine Learning Support
• Nabila Farooqi - Heroku Master
• Ruth Hinkle - Heroku Support, Documentation
• Drew McBride - Machine Learning Master
• Andrew Nehrer - Design Master