Calibrating the Predictions for Top-N Recommendations

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Sato, Masahiro
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910571932155904
author Sato, Masahiro
author_facet Sato, Masahiro
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2408_11596
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Calibrating the Predictions for Top-N Recommendations
Sato, Masahiro
Information Retrieval
Machine Learning
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.
title Calibrating the Predictions for Top-N Recommendations
topic Information Retrieval
Machine Learning
url https://arxiv.org/abs/2408.11596