Learn to rank models are such an approach they are an application of machine learning that takes the set of features and determines a relevancy score. In general, most forms of machine learning can be adapted to create LETOR models, but the particular application in consideration here is a supervised machine learning model. The data comes from CSV files and represents search queries from Home Depot’s website. The objective is to predict relevancy scores for given products from their titles from the user’s search terms. This is an involved multi-step process that is elaborated on over the course of this analysis. The models implemented here are all done in Python 3. Anaconda Python was used as the underlying Python distribution and the code was written via a Jupyter notebook.
Learning To Rank Project –3/17/19
Published by
