Layered Evaluation for Data Discovery and Recommendation Systems: An Initial Set of Principles
Nikos Manouselis, Charalampos Karagiannidis, Demetrios G. Sampson · 2014
This paper examines how a layered evaluation framework proposed for adaptive systems (AS) can be applied in the case of recommender systems (RecSys). Our analysis indicates that implementing a layered-based evaluation has the potential to facilitate a more detailed and informed evaluation of RecSys, allowing researchers and developers to better understand how to improve them.