Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Merkov, Alexander, Rohde, David, Gilotte, Alexandre, Heymann, Benjamin
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2508.10479
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916938709467136
author Merkov, Alexander
Rohde, David
Gilotte, Alexandre
Heymann, Benjamin
author_facet Merkov, Alexander
Rohde, David
Gilotte, Alexandre
Heymann, Benjamin
contents Unobserved confounding arises when an unmeasured feature influences both the treatment and the outcome, leading to biased causal effect estimates. This issue undermines observational studies in fields like economics, medicine, ecology or epidemiology. Recommender systems leveraging fully observed data seem not to be vulnerable to this problem. However many standard practices in recommender systems result in observed features being ignored, resulting in effectively the same problem. This paper will show that numerous common practices such as feature engineering, A/B testing and modularization can in fact introduce confounding into recommendation systems and hamper their performance. Several illustrations of the phenomena are provided, supported by simulation studies with practical suggestions about how practitioners may reduce or avoid the affects of confounding in real systems.
format Preprint
id arxiv_https___arxiv_org_abs_2508_10479
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Confounding is a Pervasive Problem in Real World Recommender Systems
Merkov, Alexander
Rohde, David
Gilotte, Alexandre
Heymann, Benjamin
Machine Learning
Information Retrieval
Unobserved confounding arises when an unmeasured feature influences both the treatment and the outcome, leading to biased causal effect estimates. This issue undermines observational studies in fields like economics, medicine, ecology or epidemiology. Recommender systems leveraging fully observed data seem not to be vulnerable to this problem. However many standard practices in recommender systems result in observed features being ignored, resulting in effectively the same problem. This paper will show that numerous common practices such as feature engineering, A/B testing and modularization can in fact introduce confounding into recommendation systems and hamper their performance. Several illustrations of the phenomena are provided, supported by simulation studies with practical suggestions about how practitioners may reduce or avoid the affects of confounding in real systems.
title Confounding is a Pervasive Problem in Real World Recommender Systems
topic Machine Learning
Information Retrieval
url https://arxiv.org/abs/2508.10479