Temporal heterogeneous interaction graph embedding for next-item recommendation

In the scenario of next-item recommendation, previous methods attempt to model user preferences by capturing the evolution of sequential interactions. However, their sequential expression is often limited, without modeling complex dynamics that short-term demands can often be influenced by long-term...

Full description

Saved in:
Bibliographic Details
Main Authors: JI, Yugang, YIN, Mingyang, FANG, Yuan, YANG, Hongxia, WANG, Xiangwei, JIA, Tianrui, SHI, Chuan
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2020
Subjects:
Online Access:https://ink.library.smu.edu.sg/sis_research/5157
https://ink.library.smu.edu.sg/context/sis_research/article/6160/viewcontent/ECML20_THIGE.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-6160
record_format dspace
spelling sg-smu-ink.sis_research-61602023-08-02T09:04:12Z Temporal heterogeneous interaction graph embedding for next-item recommendation JI, Yugang YIN, Mingyang FANG, Yuan YANG, Hongxia WANG, Xiangwei JIA, Tianrui SHI, Chuan In the scenario of next-item recommendation, previous methods attempt to model user preferences by capturing the evolution of sequential interactions. However, their sequential expression is often limited, without modeling complex dynamics that short-term demands can often be influenced by long-term habits. Moreover, few of them take into account the heterogeneous types of interaction between users and items. In this paper, we model such complex data as a Temporal Heterogeneous Interaction Graph (THIG) and learn both user and item embeddings on THIGs to address next-item recommendation. The main challenges involve two aspects: the complex dynamics and rich heterogeneity of interactions. We propose THIG Embedding (THIGE) which models the complex dynamics so that evolving short-term demands are guided by long-term historical habits, and leverages the rich heterogeneity to express the latent relevance of different-typed preferences. Extensive experiments on real-world datasets demonstrate that THIGE consistently outperforms the state-of-the-art methods. 2020-09-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/5157 info:doi/10.1007/978-3-030-67664-3_19 https://ink.library.smu.edu.sg/context/sis_research/article/6160/viewcontent/ECML20_THIGE.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 Temporal heterogeneous interaction graph Next-item recommendation Short-term demands Long-term habit Computer Engineering Databases and Information Systems
institution Singapore Management University
building SMU Libraries
continent Asia
country Singapore
Singapore
content_provider SMU Libraries
collection InK@SMU
language English
topic Temporal heterogeneous interaction graph
Next-item recommendation
Short-term demands
Long-term habit
Computer Engineering
Databases and Information Systems
spellingShingle Temporal heterogeneous interaction graph
Next-item recommendation
Short-term demands
Long-term habit
Computer Engineering
Databases and Information Systems
JI, Yugang
YIN, Mingyang
FANG, Yuan
YANG, Hongxia
WANG, Xiangwei
JIA, Tianrui
SHI, Chuan
Temporal heterogeneous interaction graph embedding for next-item recommendation
description In the scenario of next-item recommendation, previous methods attempt to model user preferences by capturing the evolution of sequential interactions. However, their sequential expression is often limited, without modeling complex dynamics that short-term demands can often be influenced by long-term habits. Moreover, few of them take into account the heterogeneous types of interaction between users and items. In this paper, we model such complex data as a Temporal Heterogeneous Interaction Graph (THIG) and learn both user and item embeddings on THIGs to address next-item recommendation. The main challenges involve two aspects: the complex dynamics and rich heterogeneity of interactions. We propose THIG Embedding (THIGE) which models the complex dynamics so that evolving short-term demands are guided by long-term historical habits, and leverages the rich heterogeneity to express the latent relevance of different-typed preferences. Extensive experiments on real-world datasets demonstrate that THIGE consistently outperforms the state-of-the-art methods.
format text
author JI, Yugang
YIN, Mingyang
FANG, Yuan
YANG, Hongxia
WANG, Xiangwei
JIA, Tianrui
SHI, Chuan
author_facet JI, Yugang
YIN, Mingyang
FANG, Yuan
YANG, Hongxia
WANG, Xiangwei
JIA, Tianrui
SHI, Chuan
author_sort JI, Yugang
title Temporal heterogeneous interaction graph embedding for next-item recommendation
title_short Temporal heterogeneous interaction graph embedding for next-item recommendation
title_full Temporal heterogeneous interaction graph embedding for next-item recommendation
title_fullStr Temporal heterogeneous interaction graph embedding for next-item recommendation
title_full_unstemmed Temporal heterogeneous interaction graph embedding for next-item recommendation
title_sort temporal heterogeneous interaction graph embedding for next-item recommendation
publisher Institutional Knowledge at Singapore Management University
publishDate 2020
url https://ink.library.smu.edu.sg/sis_research/5157
https://ink.library.smu.edu.sg/context/sis_research/article/6160/viewcontent/ECML20_THIGE.pdf
_version_ 1773551428787240960