Investigating the Limits of Graph Foundation Model in Real-World Travel Recommendation Systems

Nayoung Lee; Gunmin Lee; Donghun Lee

doi:10.1007/978-981-96-8197-6_13

Investigating the Limits of Graph Foundation Model in Real-World Travel Recommendation Systems

Nayoung Lee
, Gunmin Lee
, Donghun Lee^*

^*Corresponding author for this work

Department of Physics

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Graph foundation models (GFMs) have demonstrated remarkable potential in capturing intricate relational patterns, achieving state-of-the-art results in numerous graph-centric tasks. However, their real-world applicability remains underexplored in highly domain-specific contexts, such as travel recommendation. In this paper, we present a comprehensive evaluation of GFMs for large-scale travel recommendation tasks using a bipartite user–destination dataset of 86,761 travelers within South Korea. We compare representative GFM against both conventional graph-based methods and vector-based methods. Contrary to the prevailing expectation that GFMs should outperform traditional architectures, our empirical findings reveal that domain-specific constraints can dilute the benefits of extensive multi-hop message passing, leading to suboptimal performance. Our work highlights a critical need to validate GFMs against domain-specific constraints, offering a roadmap for their future adaptation and optimization in real-world applications.

Original language	English
Title of host publication	Trends and Applications in Knowledge Discovery and Data Mining - PAKDD 2025 Workshops, ADUR, FairPC, GLFM, PM4B and RAFDA, Sydney, NSW, Australia, June 10–13, 2025, Proceedings
Editors	Shuhan Yuan, Fragkiskos Malliaros, Xin Zheng
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	174-185
Number of pages	12
ISBN (Print)	9789819681969
DOIs	https://doi.org/10.1007/978-981-96-8197-6_13
Publication status	Published - 2025
Event	Workshop on Advanced Data-Driven Techniques for Urban Resilience, ADUR 2025, Workshop of Foundational AI for Pervasive Computing, FairPC 2025, Workshop on Graph Learning with Foundation Models, GLFM 2025, Workshop on Pattern Mining and Machine Learning for Bioinformatics, PM4B 2025, Workshop on Research and Applications of Foundation Models for Data Mining and Affective Computing, RAFDA 2025, held in conjunction with the 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025 - Sydney, Australia Duration: 2025 Jun 10 → 2025 Jun 13

Publication series

Name	Lecture Notes in Computer Science
Volume	15835 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	Workshop on Advanced Data-Driven Techniques for Urban Resilience, ADUR 2025, Workshop of Foundational AI for Pervasive Computing, FairPC 2025, Workshop on Graph Learning with Foundation Models, GLFM 2025, Workshop on Pattern Mining and Machine Learning for Bioinformatics, PM4B 2025, Workshop on Research and Applications of Foundation Models for Data Mining and Affective Computing, RAFDA 2025, held in conjunction with the 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025
Country/Territory	Australia
City	Sydney
Period	25/6/10 → 25/6/13

Bibliographical note

Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

Keywords

Domain-specific constraints
Graph foundation model
Travel recommendation

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-981-96-8197-6_13

Cite this

Lee, N., Lee, G., & Lee, D. (2025). Investigating the Limits of Graph Foundation Model in Real-World Travel Recommendation Systems. In S. Yuan, F. Malliaros, & X. Zheng (Eds.), Trends and Applications in Knowledge Discovery and Data Mining - PAKDD 2025 Workshops, ADUR, FairPC, GLFM, PM4B and RAFDA, Sydney, NSW, Australia, June 10–13, 2025, Proceedings (pp. 174-185). (Lecture Notes in Computer Science; Vol. 15835 LNAI). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-96-8197-6_13

Investigating the Limits of Graph Foundation Model in Real-World Travel Recommendation Systems. / Lee, Nayoung; Lee, Gunmin; Lee, Donghun.
Trends and Applications in Knowledge Discovery and Data Mining - PAKDD 2025 Workshops, ADUR, FairPC, GLFM, PM4B and RAFDA, Sydney, NSW, Australia, June 10–13, 2025, Proceedings. ed. / Shuhan Yuan; Fragkiskos Malliaros; Xin Zheng. Springer Science and Business Media Deutschland GmbH, 2025. p. 174-185 (Lecture Notes in Computer Science; Vol. 15835 LNAI).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Lee, N, Lee, G & Lee, D 2025, Investigating the Limits of Graph Foundation Model in Real-World Travel Recommendation Systems. in S Yuan, F Malliaros & X Zheng (eds), Trends and Applications in Knowledge Discovery and Data Mining - PAKDD 2025 Workshops, ADUR, FairPC, GLFM, PM4B and RAFDA, Sydney, NSW, Australia, June 10–13, 2025, Proceedings. Lecture Notes in Computer Science, vol. 15835 LNAI, Springer Science and Business Media Deutschland GmbH, pp. 174-185, Workshop on Advanced Data-Driven Techniques for Urban Resilience, ADUR 2025, Workshop of Foundational AI for Pervasive Computing, FairPC 2025, Workshop on Graph Learning with Foundation Models, GLFM 2025, Workshop on Pattern Mining and Machine Learning for Bioinformatics, PM4B 2025, Workshop on Research and Applications of Foundation Models for Data Mining and Affective Computing, RAFDA 2025, held in conjunction with the 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025, Sydney, Australia, 25/6/10. https://doi.org/10.1007/978-981-96-8197-6_13

Lee N, Lee G, Lee D. Investigating the Limits of Graph Foundation Model in Real-World Travel Recommendation Systems. In Yuan S, Malliaros F, Zheng X, editors, Trends and Applications in Knowledge Discovery and Data Mining - PAKDD 2025 Workshops, ADUR, FairPC, GLFM, PM4B and RAFDA, Sydney, NSW, Australia, June 10–13, 2025, Proceedings. Springer Science and Business Media Deutschland GmbH. 2025. p. 174-185. (Lecture Notes in Computer Science). doi: 10.1007/978-981-96-8197-6_13

Lee, Nayoung ; Lee, Gunmin ; Lee, Donghun. / Investigating the Limits of Graph Foundation Model in Real-World Travel Recommendation Systems. Trends and Applications in Knowledge Discovery and Data Mining - PAKDD 2025 Workshops, ADUR, FairPC, GLFM, PM4B and RAFDA, Sydney, NSW, Australia, June 10–13, 2025, Proceedings. editor / Shuhan Yuan ; Fragkiskos Malliaros ; Xin Zheng. Springer Science and Business Media Deutschland GmbH, 2025. pp. 174-185 (Lecture Notes in Computer Science).

@inproceedings{2d83d42efe2e4335a1a58d77f33df917,

title = "Investigating the Limits of Graph Foundation Model in Real-World Travel Recommendation Systems",

abstract = "Graph foundation models (GFMs) have demonstrated remarkable potential in capturing intricate relational patterns, achieving state-of-the-art results in numerous graph-centric tasks. However, their real-world applicability remains underexplored in highly domain-specific contexts, such as travel recommendation. In this paper, we present a comprehensive evaluation of GFMs for large-scale travel recommendation tasks using a bipartite user–destination dataset of 86,761 travelers within South Korea. We compare representative GFM against both conventional graph-based methods and vector-based methods. Contrary to the prevailing expectation that GFMs should outperform traditional architectures, our empirical findings reveal that domain-specific constraints can dilute the benefits of extensive multi-hop message passing, leading to suboptimal performance. Our work highlights a critical need to validate GFMs against domain-specific constraints, offering a roadmap for their future adaptation and optimization in real-world applications.",

keywords = "Domain-specific constraints, Graph foundation model, Travel recommendation",

author = "Nayoung Lee and Gunmin Lee and Donghun Lee",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.; Workshop on Advanced Data-Driven Techniques for Urban Resilience, ADUR 2025, Workshop of Foundational AI for Pervasive Computing, FairPC 2025, Workshop on Graph Learning with Foundation Models, GLFM 2025, Workshop on Pattern Mining and Machine Learning for Bioinformatics, PM4B 2025, Workshop on Research and Applications of Foundation Models for Data Mining and Affective Computing, RAFDA 2025, held in conjunction with the 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025 ; Conference date: 10-06-2025 Through 13-06-2025",

year = "2025",

doi = "10.1007/978-981-96-8197-6\_13",

language = "English",

isbn = "9789819681969",

series = "Lecture Notes in Computer Science",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "174--185",

editor = "Shuhan Yuan and Fragkiskos Malliaros and Xin Zheng",

booktitle = "Trends and Applications in Knowledge Discovery and Data Mining - PAKDD 2025 Workshops, ADUR, FairPC, GLFM, PM4B and RAFDA, Sydney, NSW, Australia, June 10–13, 2025, Proceedings",

address = "Germany",

}

TY - GEN

T1 - Investigating the Limits of Graph Foundation Model in Real-World Travel Recommendation Systems

AU - Lee, Nayoung

AU - Lee, Gunmin

AU - Lee, Donghun

N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

PY - 2025

Y1 - 2025

N2 - Graph foundation models (GFMs) have demonstrated remarkable potential in capturing intricate relational patterns, achieving state-of-the-art results in numerous graph-centric tasks. However, their real-world applicability remains underexplored in highly domain-specific contexts, such as travel recommendation. In this paper, we present a comprehensive evaluation of GFMs for large-scale travel recommendation tasks using a bipartite user–destination dataset of 86,761 travelers within South Korea. We compare representative GFM against both conventional graph-based methods and vector-based methods. Contrary to the prevailing expectation that GFMs should outperform traditional architectures, our empirical findings reveal that domain-specific constraints can dilute the benefits of extensive multi-hop message passing, leading to suboptimal performance. Our work highlights a critical need to validate GFMs against domain-specific constraints, offering a roadmap for their future adaptation and optimization in real-world applications.

AB - Graph foundation models (GFMs) have demonstrated remarkable potential in capturing intricate relational patterns, achieving state-of-the-art results in numerous graph-centric tasks. However, their real-world applicability remains underexplored in highly domain-specific contexts, such as travel recommendation. In this paper, we present a comprehensive evaluation of GFMs for large-scale travel recommendation tasks using a bipartite user–destination dataset of 86,761 travelers within South Korea. We compare representative GFM against both conventional graph-based methods and vector-based methods. Contrary to the prevailing expectation that GFMs should outperform traditional architectures, our empirical findings reveal that domain-specific constraints can dilute the benefits of extensive multi-hop message passing, leading to suboptimal performance. Our work highlights a critical need to validate GFMs against domain-specific constraints, offering a roadmap for their future adaptation and optimization in real-world applications.

KW - Domain-specific constraints

KW - Graph foundation model

KW - Travel recommendation

UR - https://www.scopus.com/pages/publications/105009765940

U2 - 10.1007/978-981-96-8197-6_13

DO - 10.1007/978-981-96-8197-6_13

M3 - Conference contribution

AN - SCOPUS:105009765940

SN - 9789819681969

T3 - Lecture Notes in Computer Science

SP - 174

EP - 185

BT - Trends and Applications in Knowledge Discovery and Data Mining - PAKDD 2025 Workshops, ADUR, FairPC, GLFM, PM4B and RAFDA, Sydney, NSW, Australia, June 10–13, 2025, Proceedings

A2 - Yuan, Shuhan

A2 - Malliaros, Fragkiskos

A2 - Zheng, Xin

PB - Springer Science and Business Media Deutschland GmbH

T2 - Workshop on Advanced Data-Driven Techniques for Urban Resilience, ADUR 2025, Workshop of Foundational AI for Pervasive Computing, FairPC 2025, Workshop on Graph Learning with Foundation Models, GLFM 2025, Workshop on Pattern Mining and Machine Learning for Bioinformatics, PM4B 2025, Workshop on Research and Applications of Foundation Models for Data Mining and Affective Computing, RAFDA 2025, held in conjunction with the 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025

Y2 - 10 June 2025 through 13 June 2025

ER -