Vague Preference Policy Learning for Conversational Recommendation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Zhang, Gangyi, Gao, Chongming, Lei, Wenqiang, Guo, Xiaojie, Li, Shijun, Chen, Hongshen, Ding, Zhuozhi, Xu, Sulong, Wu, Lingfei
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916760631902208
author Zhang, Gangyi
Gao, Chongming
Lei, Wenqiang
Guo, Xiaojie
Li, Shijun
Chen, Hongshen
Ding, Zhuozhi
Xu, Sulong
Wu, Lingfei
author_facet Zhang, Gangyi
Gao, Chongming
Lei, Wenqiang
Guo, Xiaojie
Li, Shijun
Chen, Hongshen
Ding, Zhuozhi
Xu, Sulong
Wu, Lingfei
contents Conversational recommendation systems (CRS) commonly assume users have clear preferences, leading to potential over-filtering of relevant alternatives. However, users often exhibit vague, non-binary preferences. We introduce the Vague Preference Multi-round Conversational Recommendation (VPMCR) scenario, employing a soft estimation mechanism to accommodate users' vague and dynamic preferences while mitigating over-filtering. In VPMCR, we propose Vague Preference Policy Learning (VPPL), consisting of Ambiguity-aware Soft Estimation (ASE) and Dynamism-aware Policy Learning (DPL). ASE captures preference vagueness by estimating scores for clicked and non-clicked options, using a choice-based approach and time-aware preference decay. DPL leverages ASE's preference distribution to guide the conversation and adapt to preference changes for recommendations or attribute queries. Extensive experiments demonstrate VPPL's effectiveness within VPMCR, outperforming existing methods and setting a new benchmark. Our work advances CRS by accommodating users' inherent ambiguity and relative decision-making processes, improving real-world applicability.
format Preprint
id arxiv_https___arxiv_org_abs_2306_04487
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Vague Preference Policy Learning for Conversational Recommendation
Zhang, Gangyi
Gao, Chongming
Lei, Wenqiang
Guo, Xiaojie
Li, Shijun
Chen, Hongshen
Ding, Zhuozhi
Xu, Sulong
Wu, Lingfei
Information Retrieval
Conversational recommendation systems (CRS) commonly assume users have clear preferences, leading to potential over-filtering of relevant alternatives. However, users often exhibit vague, non-binary preferences. We introduce the Vague Preference Multi-round Conversational Recommendation (VPMCR) scenario, employing a soft estimation mechanism to accommodate users' vague and dynamic preferences while mitigating over-filtering. In VPMCR, we propose Vague Preference Policy Learning (VPPL), consisting of Ambiguity-aware Soft Estimation (ASE) and Dynamism-aware Policy Learning (DPL). ASE captures preference vagueness by estimating scores for clicked and non-clicked options, using a choice-based approach and time-aware preference decay. DPL leverages ASE's preference distribution to guide the conversation and adapt to preference changes for recommendations or attribute queries. Extensive experiments demonstrate VPPL's effectiveness within VPMCR, outperforming existing methods and setting a new benchmark. Our work advances CRS by accommodating users' inherent ambiguity and relative decision-making processes, improving real-world applicability.
title Vague Preference Policy Learning for Conversational Recommendation
topic Information Retrieval
url https://arxiv.org/abs/2306.04487