Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wang, Xinyan, Liu, Jinshuo, Xie, Kaijian, Wang, Meng, Bi, Cheng, Deng, Juan, Pan, Jeff
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2503.08189
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914010139459584
author Wang, Xinyan
Liu, Jinshuo
Xie, Kaijian
Wang, Meng
Bi, Cheng
Deng, Juan
Pan, Jeff
author_facet Wang, Xinyan
Liu, Jinshuo
Xie, Kaijian
Wang, Meng
Bi, Cheng
Deng, Juan
Pan, Jeff
contents Current Continual Knowledge Graph Embedding (CKGE) methods primarily rely on translation-based embedding approaches, leveraging previously acquired knowledge to initialize new facts. While these methods often integrate fine-tuning or continual learning strategies to enhance efficiency, they compromise prediction accuracy and lack support for complex relational structures (e.g., multi-hop relations). To address these limitations, we propose STCKGE, a novel CKGE framework based on spatial transformation. In this framework, entity positions are jointly determined by base position vectors and offset vectors, enabling the model to represent complex relations more effectively while supporting efficient embedding updates for both new and existing knowledge through simple spatial operations, without relying on traditional continual learning techniques. Furthermore, we introduce a bidirectional collaborative update strategy and a balanced embedding method to guide parameter updates, effectively minimizing training costs while improving model accuracy. We comprehensively evaluate our model on seven public datasets and a newly constructed dataset (MULTI) focusing on multi-hop relationships. Experimental results confirm STCKGE's strong performance in multi-hop relationship learning and prediction accuracy, with an average MRR improvement of 5.4\%.
format Preprint
id arxiv_https___arxiv_org_abs_2503_08189
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle STCKGE:Continual Knowledge Graph Embedding Based on Spatial Transformation
Wang, Xinyan
Liu, Jinshuo
Xie, Kaijian
Wang, Meng
Bi, Cheng
Deng, Juan
Pan, Jeff
Information Retrieval
68T30
E.2
Current Continual Knowledge Graph Embedding (CKGE) methods primarily rely on translation-based embedding approaches, leveraging previously acquired knowledge to initialize new facts. While these methods often integrate fine-tuning or continual learning strategies to enhance efficiency, they compromise prediction accuracy and lack support for complex relational structures (e.g., multi-hop relations). To address these limitations, we propose STCKGE, a novel CKGE framework based on spatial transformation. In this framework, entity positions are jointly determined by base position vectors and offset vectors, enabling the model to represent complex relations more effectively while supporting efficient embedding updates for both new and existing knowledge through simple spatial operations, without relying on traditional continual learning techniques. Furthermore, we introduce a bidirectional collaborative update strategy and a balanced embedding method to guide parameter updates, effectively minimizing training costs while improving model accuracy. We comprehensively evaluate our model on seven public datasets and a newly constructed dataset (MULTI) focusing on multi-hop relationships. Experimental results confirm STCKGE's strong performance in multi-hop relationship learning and prediction accuracy, with an average MRR improvement of 5.4\%.
title STCKGE:Continual Knowledge Graph Embedding Based on Spatial Transformation
topic Information Retrieval
68T30
E.2
url https://arxiv.org/abs/2503.08189