We will explore techniques to determine the amount of similarity between documents. Specifically we will look at the intuition behind tf-idf and cosine similarity. With that as a foundation we will see how to compute these metrics with the natural language tool kit.