This document describes a Yelp data challenge to predict user ratings of businesses from user review text using classical machine learning algorithms and deep learning techniques. It provides details on the problem definition, preprocessing steps, models used, and results. Classical machine learning approaches such as Naive Bayes, logistic regression, and SVM were able to predict ratings with around 67% accuracy, while a convolutional neural network achieved slightly higher accuracy of 73.5%. A Docker image containing the code was also created to allow easy running of the models.