This is a project that began as part of an internal Machine Learning Multidisciplinary Hackathon where the objective was to adapt the Spotify dataset on Worldwide Daily Song Ranking (available on kaggle) to a Learning to Rank task.
The repository contains the code to set up and build a Learning to Rank (LTR) system starting from the available data, creating and manipulating the training set and the test set, and then training a ranking model using open source libraries.
More details about the project are available in the following blog posts:
part1 - https://sease.io/2020/12/a-learning-to-rank-project-on-a-daily-song-ranking-problem.html
part2 - https://sease.io/2021/02/a-learning-to-rank-project-on-a-daily-song-ranking-problem-part-2.html
part3 - https://sease.io/2021/03/a-learning-to-rank-project-on-a-daily-song-ranking-problem-part-3.html
part4 - https://sease.io/2021/05/a-learning-to-rank-project-on-a-daily-song-ranking-problem-part-4.html
For simplicity, the code has been divided into 4 parts based on the topic of each blog post and can be found in the namesake folders. However, each file can be adapted and reused depending on the purpose.