Towards a Technical Debt for Recommender System

Balancing the management of technical debt within recommender systems requires effectively juggling the introduction of new features with the ongoing maintenance and enhancement of the current system. Within the realm of recommender systems, technical debt encompasses the trade-offs and expedient choices made during the development and upkeep of the recommendation system, which could potentially have adverse effects on its long-term performance, scalability, and maintainability. In this vision paper, our objective is to kickstart a research direction regarding Technical Debt in Recommender Systems. We identified 15 potential factors, along with detailed explanations outlining why it is advisable to consider them.