The aim of this task is to identify the relevant, high-quality web pages from a pool of user-curated web pages, for the identification of “ad-worthy” web pages. The challenge requires to build large-scale, end-to-end machine learning models that can classify a website as either “relevant” or “irrelevant”, based on attributes such as alchemy category and its score, meta-information of the web pages and a one-line description of the content of each webpage. This task aims to gently introduce you to the domain of NLP, as you would be required to convert the string attributes of the dataset to some form of numerical data, and then construct your ML models on this numerical data.