A Comprehensive Review of User Interaction for Recommendation Systems

Abstract

Listen

Translate article

In recent years, Recommendation Systems have become integral to the online experiences of consumers, particularly those that effectively integrate user interactions into their algorithms, enhancing both efficiency and adaptability. This article presents a comprehensive systematic review of the literature addressing classical problems in recommendation systems, specifically focusing on the consideration of user interaction. We employed a rigorous systematic literature review methodology, critically analyzing various proposals to identify their limitations, characteristics, and potential avenues for further research. Our investigation involved mapping relevant studies that examine how user interaction with recommendation systems is addressed and determining the extent to which this aspect has been explored. We established strict inclusion and exclusion criteria to select academic publications, resulting in a curated set of 29 scientific papers. The findings offer a snapshot of the primary characteristics of the identified works, revealing significant gaps that can inform future research directions. Our analysis indicates that most studies addressing user interaction emphasize preference elicitation and feedback mechanisms, predominantly focusing on improving the accuracy of recommendation rankings, with a notable concentration on the e-commerce domain.