Entity class-restricted twin attention-based aggregation for geographic knowledge graph inference

Abstract Geographic knowledge graphs (GeoKGs) describe and simulate geographic phenomena by storing geographic entities with their semantic relations, providing a foundation for many real-world tasks such as urban function identification. However, GeoKGs are often incomplete due to manual constructi...

Full description

Saved in:
Bibliographic Details
Main Authors: Jinlan Kong, Xiaohui He, Haichuan Fang, Roujia Ji, Peng Yang, Shuang Li, Haofei Li, Haonan Sun, Mengjia Qiao, Panle Li, Xijie Cheng, Qinglei Zhou, Jiandong Shang
Format: Article
Language:English
Published: Springer 2025-08-01
Series:Complex & Intelligent Systems
Subjects:
Online Access:https://doi.org/10.1007/s40747-025-02060-y
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849225743948578816
author Jinlan Kong
Xiaohui He
Haichuan Fang
Roujia Ji
Peng Yang
Shuang Li
Haofei Li
Haonan Sun
Mengjia Qiao
Panle Li
Xijie Cheng
Qinglei Zhou
Jiandong Shang
author_facet Jinlan Kong
Xiaohui He
Haichuan Fang
Roujia Ji
Peng Yang
Shuang Li
Haofei Li
Haonan Sun
Mengjia Qiao
Panle Li
Xijie Cheng
Qinglei Zhou
Jiandong Shang
author_sort Jinlan Kong
collection DOAJ
description Abstract Geographic knowledge graphs (GeoKGs) describe and simulate geographic phenomena by storing geographic entities with their semantic relations, providing a foundation for many real-world tasks such as urban function identification. However, GeoKGs are often incomplete due to manual construction, leading to inaccurate and/or suboptimal decision-making. This situation requires GeoKG inference (GeoKGI), which involves learning new knowledge by projecting entities and relations into embedding spaces to infer missing facts and complete GeoKGs. Nevertheless, current inference methods mainly focus on modeling universal knowledge graphs (KGs), which cannot be directly transferred into GeoKGs as, unlike KGs, their entities contain specific class information. Inference methods on universal KGs lack a clear manner of modeling such entity class information. Moreover, they often ignore the representation learning of relations, resulting in inadequate modeling of GeoKGs. To address this issue, we propose a novel entity class-restricted twin attention-based aggregation (ECTAA) framework for GeoKGI. Briefly, we first restricted entities by encapsulating the class information within their vector representations to aid the model in better understanding and distinguishing between different entities and their attributes. We then designed a twin aggregation module with attention mechanisms to adaptively incorporate context information into both entity and relation representations. This module simultaneously considers the impact of context on entities and relations, further promoting the capture of GeoKG structural information. Experimental results on five GeoKG datasets validated the effectiveness of ECTAA compared to several state-of-the-art universal KG inference methods in predicting geographic facts.
format Article
id doaj-art-12da82eff77a44bc8d7c6a5413fcebcd
institution Kabale University
issn 2199-4536
2198-6053
language English
publishDate 2025-08-01
publisher Springer
record_format Article
series Complex & Intelligent Systems
spelling doaj-art-12da82eff77a44bc8d7c6a5413fcebcd2025-08-24T12:02:21ZengSpringerComplex & Intelligent Systems2199-45362198-60532025-08-01111011910.1007/s40747-025-02060-yEntity class-restricted twin attention-based aggregation for geographic knowledge graph inferenceJinlan Kong0Xiaohui He1Haichuan Fang2Roujia Ji3Peng Yang4Shuang Li5Haofei Li6Haonan Sun7Mengjia Qiao8Panle Li9Xijie Cheng10Qinglei Zhou11Jiandong Shang12School of Computer Science and Artificial Intelligence, Zhengzhou UniversitySchool of Geoscience and Technology, Zhengzhou UniversitySchool of Computer Science and Artificial Intelligence, Zhengzhou UniversitySchool of Geoscience and Technology, Zhengzhou UniversitySchool of Geoscience and Technology, Zhengzhou UniversitySchool of Geoscience and Technology, Zhengzhou UniversitySchool of Computer Science and Artificial Intelligence, Zhengzhou UniversitySchool of Computer Science and Artificial Intelligence, Zhengzhou UniversitySchool of Geoscience and Technology, Zhengzhou UniversitySchool of Geoscience and Technology, Zhengzhou UniversitySchool of Geoscience and Technology, Zhengzhou UniversitySchool of Computer Science and Artificial Intelligence, Zhengzhou UniversityNational Supercomputing Center in Zhengzhou, Zhengzhou UniversityAbstract Geographic knowledge graphs (GeoKGs) describe and simulate geographic phenomena by storing geographic entities with their semantic relations, providing a foundation for many real-world tasks such as urban function identification. However, GeoKGs are often incomplete due to manual construction, leading to inaccurate and/or suboptimal decision-making. This situation requires GeoKG inference (GeoKGI), which involves learning new knowledge by projecting entities and relations into embedding spaces to infer missing facts and complete GeoKGs. Nevertheless, current inference methods mainly focus on modeling universal knowledge graphs (KGs), which cannot be directly transferred into GeoKGs as, unlike KGs, their entities contain specific class information. Inference methods on universal KGs lack a clear manner of modeling such entity class information. Moreover, they often ignore the representation learning of relations, resulting in inadequate modeling of GeoKGs. To address this issue, we propose a novel entity class-restricted twin attention-based aggregation (ECTAA) framework for GeoKGI. Briefly, we first restricted entities by encapsulating the class information within their vector representations to aid the model in better understanding and distinguishing between different entities and their attributes. We then designed a twin aggregation module with attention mechanisms to adaptively incorporate context information into both entity and relation representations. This module simultaneously considers the impact of context on entities and relations, further promoting the capture of GeoKG structural information. Experimental results on five GeoKG datasets validated the effectiveness of ECTAA compared to several state-of-the-art universal KG inference methods in predicting geographic facts.https://doi.org/10.1007/s40747-025-02060-yGeographic knowledge graphGeographic knowledge graph inferenceEntity classTwin aggregationAttention mechanism
spellingShingle Jinlan Kong
Xiaohui He
Haichuan Fang
Roujia Ji
Peng Yang
Shuang Li
Haofei Li
Haonan Sun
Mengjia Qiao
Panle Li
Xijie Cheng
Qinglei Zhou
Jiandong Shang
Entity class-restricted twin attention-based aggregation for geographic knowledge graph inference
Complex & Intelligent Systems
Geographic knowledge graph
Geographic knowledge graph inference
Entity class
Twin aggregation
Attention mechanism
title Entity class-restricted twin attention-based aggregation for geographic knowledge graph inference
title_full Entity class-restricted twin attention-based aggregation for geographic knowledge graph inference
title_fullStr Entity class-restricted twin attention-based aggregation for geographic knowledge graph inference
title_full_unstemmed Entity class-restricted twin attention-based aggregation for geographic knowledge graph inference
title_short Entity class-restricted twin attention-based aggregation for geographic knowledge graph inference
title_sort entity class restricted twin attention based aggregation for geographic knowledge graph inference
topic Geographic knowledge graph
Geographic knowledge graph inference
Entity class
Twin aggregation
Attention mechanism
url https://doi.org/10.1007/s40747-025-02060-y
work_keys_str_mv AT jinlankong entityclassrestrictedtwinattentionbasedaggregationforgeographicknowledgegraphinference
AT xiaohuihe entityclassrestrictedtwinattentionbasedaggregationforgeographicknowledgegraphinference
AT haichuanfang entityclassrestrictedtwinattentionbasedaggregationforgeographicknowledgegraphinference
AT roujiaji entityclassrestrictedtwinattentionbasedaggregationforgeographicknowledgegraphinference
AT pengyang entityclassrestrictedtwinattentionbasedaggregationforgeographicknowledgegraphinference
AT shuangli entityclassrestrictedtwinattentionbasedaggregationforgeographicknowledgegraphinference
AT haofeili entityclassrestrictedtwinattentionbasedaggregationforgeographicknowledgegraphinference
AT haonansun entityclassrestrictedtwinattentionbasedaggregationforgeographicknowledgegraphinference
AT mengjiaqiao entityclassrestrictedtwinattentionbasedaggregationforgeographicknowledgegraphinference
AT panleli entityclassrestrictedtwinattentionbasedaggregationforgeographicknowledgegraphinference
AT xijiecheng entityclassrestrictedtwinattentionbasedaggregationforgeographicknowledgegraphinference
AT qingleizhou entityclassrestrictedtwinattentionbasedaggregationforgeographicknowledgegraphinference
AT jiandongshang entityclassrestrictedtwinattentionbasedaggregationforgeographicknowledgegraphinference