Factorization Machines for Recommendation Systems

As a Data Scientist that works on Feed Personalization, I find it it important to stay up to date with the current state of Machine Learning and its applications. Most of the time, using some of the better-known recommendation algorithms yields good initial results; however, sometimes a change in the model is essential to provide customers with that extra boost that helps increase engagement in their apps. This is how we ended up reading and researching the use of Factorization Machines (FM) to improve our personalization engine. This blogpost will provide brief explanation of Factorization Machines (FM) and their applications to the cold-start recommendation problem. FM models are at the cutting edge of Machine Learning techniques for personalization; they have proven to be an extremely powerful tool with enough expressive capacity to generalize methods such as Matrix/Tensor Factorization and Polynomial Kernel regression.