Estimating propensity for causality-based recommendation without exposure data
Causality-based recommendation systems focus on the causal effects of user-item interactions resulting from item exposure (i.e., which items are recommended or exposed to the user), as opposed to conventional correlation-based recommendation. They are gaining popularity due to their multi-sided bene...
Saved in:
Main Authors: | , , |
---|---|
Format: | text |
Language: | English |
Published: |
Institutional Knowledge at Singapore Management University
2023
|
Subjects: | |
Online Access: | https://ink.library.smu.edu.sg/sis_research/8250 https://ink.library.smu.edu.sg/context/sis_research/article/9253/viewcontent/LZZ_NIPS_Causal_2.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Singapore Management University |
Language: | English |
id |
sg-smu-ink.sis_research-9253 |
---|---|
record_format |
dspace |
spelling |
sg-smu-ink.sis_research-92532024-09-20T07:06:17Z Estimating propensity for causality-based recommendation without exposure data LIU, Zhongzhou FANG, Yuan WU, Min Causality-based recommendation systems focus on the causal effects of user-item interactions resulting from item exposure (i.e., which items are recommended or exposed to the user), as opposed to conventional correlation-based recommendation. They are gaining popularity due to their multi-sided benefits to users, sellers and platforms alike. However, existing causality-based recommendation methods require additional input in the form of exposure data and/or propensity scores (i.e., the probability of exposure) for training. Such data, crucial for modeling causality in recommendation, are often not available in real-world situations due to technical or privacy constraints. In this paper, we bridge the gap by proposing a new framework, called Propensity Estimation for Causality-based Recommendation (PropCare). It can estimate the propensity and exposure from a more practical setup, where only interaction data are available without any ground truth on exposure or propensity in training and inference. We demonstrate that, by relating the pairwise characteristics between propensity and item popularity, PropCare enables competitive causality-based recommendation given only the conventional interaction data. We further present a theoretical analysis on the bias of the causal effect under our model estimation. Finally, we empirically evaluate PropCare through both quantitative and qualitative experiments. 2023-12-01T08:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/8250 https://ink.library.smu.edu.sg/context/sis_research/article/9253/viewcontent/LZZ_NIPS_Causal_2.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Causality-based recommendation Exposure Propensity estimation Artificial Intelligence and Robotics |
institution |
Singapore Management University |
building |
SMU Libraries |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
SMU Libraries |
collection |
InK@SMU |
language |
English |
topic |
Causality-based recommendation Exposure Propensity estimation Artificial Intelligence and Robotics |
spellingShingle |
Causality-based recommendation Exposure Propensity estimation Artificial Intelligence and Robotics LIU, Zhongzhou FANG, Yuan WU, Min Estimating propensity for causality-based recommendation without exposure data |
description |
Causality-based recommendation systems focus on the causal effects of user-item interactions resulting from item exposure (i.e., which items are recommended or exposed to the user), as opposed to conventional correlation-based recommendation. They are gaining popularity due to their multi-sided benefits to users, sellers and platforms alike. However, existing causality-based recommendation methods require additional input in the form of exposure data and/or propensity scores (i.e., the probability of exposure) for training. Such data, crucial for modeling causality in recommendation, are often not available in real-world situations due to technical or privacy constraints. In this paper, we bridge the gap by proposing a new framework, called Propensity Estimation for Causality-based Recommendation (PropCare). It can estimate the propensity and exposure from a more practical setup, where only interaction data are available without any ground truth on exposure or propensity in training and inference. We demonstrate that, by relating the pairwise characteristics between propensity and item popularity, PropCare enables competitive causality-based recommendation given only the conventional interaction data. We further present a theoretical analysis on the bias of the causal effect under our model estimation. Finally, we empirically evaluate PropCare through both quantitative and qualitative experiments. |
format |
text |
author |
LIU, Zhongzhou FANG, Yuan WU, Min |
author_facet |
LIU, Zhongzhou FANG, Yuan WU, Min |
author_sort |
LIU, Zhongzhou |
title |
Estimating propensity for causality-based recommendation without exposure data |
title_short |
Estimating propensity for causality-based recommendation without exposure data |
title_full |
Estimating propensity for causality-based recommendation without exposure data |
title_fullStr |
Estimating propensity for causality-based recommendation without exposure data |
title_full_unstemmed |
Estimating propensity for causality-based recommendation without exposure data |
title_sort |
estimating propensity for causality-based recommendation without exposure data |
publisher |
Institutional Knowledge at Singapore Management University |
publishDate |
2023 |
url |
https://ink.library.smu.edu.sg/sis_research/8250 https://ink.library.smu.edu.sg/context/sis_research/article/9253/viewcontent/LZZ_NIPS_Causal_2.pdf |
_version_ |
1814047877638389760 |