Discussion of the challenges of automating the development of recommender systems.

What is it about?

In this work, I discuss the primary challenges for the automation of recommender system development. Recommender system development is a laborious process that requires plenty of domain knowledge. I seek ways to ease the development process by automating difficult design decisions. I present a proof-of-concept system in the paper and outline my plans for the future.

Featured Image

Why is it important?

This work describes my Ph.D. research goals and problems. Automation is essential because it saves time, increases performance, and reduces the environmental impact of development. In traditional machine learning, automation is already used effectively. I seek to understand how to transfer the advantages to recommender systems.