Calibrated Recommendations: Survey and Future Directions

TL;DR

This survey reviews recent research on calibrated recommendations, which aim to align suggested items with users' past preferences, addressing diversity, bias, and fairness, while discussing technical approaches, effectiveness, limitations, and future directions.

Contribution

It provides a comprehensive overview of calibration techniques in recommender systems, highlighting recent developments, empirical findings, and practical challenges.

Findings

Calibration improves diversity and fairness in recommendations.

Empirical studies show mixed effectiveness of calibration methods.

Implementation challenges include data bias and computational complexity.

Abstract

The idea of calibrated recommendations is that the properties of the items that are suggested to users should match the distribution of their individual past preferences. Calibration techniques are therefore helpful to ensure that the recommendations provided to a user are not limited to a certain subset of the user's interests. Over the past few years, we have observed an increasing number of research works that use calibration for different purposes, including questions of diversity, biases, and fairness. In this work, we provide a survey on the recent developments in the area of calibrated recommendations. We both review existing technical approaches for calibration and provide an overview on empirical and analytical studies on the effectiveness of calibration for different use cases. Furthermore, we discuss limitations and common challenges when implementing calibration in practice.