English

Calibrating the Predictions for Top-N Recommendations

Information Retrieval 2024-08-22 v1 Machine Learning

Abstract

Well-calibrated predictions of user preferences are essential for many applications. Since recommender systems typically select the top-N items for users, calibration for those top-N items, rather than for all items, is important. We show that previous calibration methods result in miscalibrated predictions for the top-N items, despite their excellent calibration performance when evaluated on all items. In this work, we address the miscalibration in the top-N recommended items. We first define evaluation metrics for this objective and then propose a generic method to optimize calibration models focusing on the top-N items. It groups the top-N items by their ranks and optimizes distinct calibration models for each group with rank-dependent training weights. We verify the effectiveness of the proposed method for both explicit and implicit feedback datasets, using diverse classes of recommender models.

Keywords

Cite

@article{arxiv.2408.11596,
  title  = {Calibrating the Predictions for Top-N Recommendations},
  author = {Masahiro Sato},
  journal= {arXiv preprint arXiv:2408.11596},
  year   = {2024}
}

Comments

accepted at RecSys 2024

R2 v1 2026-06-28T18:19:27.530Z