SeSICL: Semantic and Structural Integrated Contrastive Learning for Knowledge Graph Error Detection

被引:1
|
作者
Liu, Xingyu [1 ]
Tang, Jielong [2 ]
Li, Mengyang [3 ]
Han, Junmei [4 ]
Xiao, Gang [4 ]
Jiang, Jianchun [3 ]
机构
[1] Chinese Acad Sci, Inst Software, State Key Lab Intelligent Game, Beijing 100190, Peoples R China
[2] Sun Yat Sen Univ, Sch Artificial Intelligence, Guangzhou 510330, Peoples R China
[3] Chinese Acad Sci, Inst Software, Integrat Innovat Ctr, Beijing 100190, Peoples R China
[4] Syst Engn Inst, Syst Res Dept, Natl Key Lab Complex Syst Simulat, AMS, Beijing 100192, Peoples R China
关键词
Knowledge graph; error detection; semantic embedding; structural embedding; contrastive learning; QUALITY;
D O I
10.1109/ACCESS.2024.3384543
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
As knowledge graphs (KGs) become more widely used in various applications, error detection for KGs has received more attention, which can reduce quality issues such as errors and inconsistencies. With the development of representation learning, embedding-based methods have significantly improved error detection performance. The recent error detection algorithm uses KG structural embedding loss and constructs a reasonable score function, ranking the confidence scores for each triplet. However, these methods ignore the factual semantics of the triplet itself, which primarily exist in the entities and relations descriptions text. Therefore, we propose Semantic and Structural Integrated Contrastive Learning(SeSICL) to simultaneously capture graph structural patterns and deep semantic features from descriptions text. Our method is based on contrastive learning without data augmentation, which utilizes encoder perturbations to generate contrasting views, making SeSICL highly suitable for complex error detection tasks and robust against real-world noise. We evaluate SeSICL on three baseline datasets with abnormal data and fluctuations. SeSICL outperforms the previous state-of-the-art methods, demonstrating our method's performance and robustness in more complex scenarios.
引用
收藏
页码:56088 / 56096
页数:9
相关论文
共 50 条
  • [21] TCKGE: Transformers with contrastive learning for knowledge graph embedding
    Xiaowei Zhang
    Quan Fang
    Jun Hu
    Shengsheng Qian
    Changsheng Xu
    International Journal of Multimedia Information Retrieval, 2022, 11 : 589 - 597
  • [22] Completion of Temporal Knowledge Graph for Historical Contrastive Learning
    Xu, Zhihong
    Qiu, Penglin
    Wang, Liqin
    Dong, Yongfeng
    Computer Engineering and Applications, 2024, 60 (22) : 154 - 161
  • [23] Contrastive Predictive Embedding for learning and inference in knowledge graph
    Liu, Chen
    Wei, Zihan
    Zhou, Lixin
    KNOWLEDGE-BASED SYSTEMS, 2025, 307
  • [24] Temporal Knowledge Graph Reasoning with Historical Contrastive Learning
    Xu, Yi
    Ou, Junjie
    Xu, Hui
    Fu, Luoyi
    THIRTY-SEVENTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, VOL 37 NO 4, 2023, : 4765 - 4773
  • [25] Federated Graph Semantic and Structural Learning
    Huang, Wenke
    Wan, Guancheng
    Ye, Mang
    Du, Bo
    IJCAI International Joint Conference on Artificial Intelligence, 2023, 2023-August : 3830 - 3838
  • [26] Federated Graph Semantic and Structural Learning
    Huang, Wenke
    Wan, Guancheng
    Ye, Mang
    Du, Bo
    PROCEEDINGS OF THE THIRTY-SECOND INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, IJCAI 2023, 2023, : 3830 - 3838
  • [27] Fraud detection in multi-relation graph: Contrastive Learning on Feature and Structural Levels
    Tang, Jiangnan
    Gu, Huanhuan
    Vukovic, Darko B.
    Xu, Guandong
    Wang, Youquan
    Tao, Haicheng
    Cao, Jie
    NEUROCOMPUTING, 2025, 637
  • [28] Knowledge Error Detection via Textual and Structural Joint Learning
    Wang, Xiaoyu
    Ao, Xiang
    Zhang, Fuwei
    Zhang, Zhao
    He, Qing
    BIG DATA MINING AND ANALYTICS, 2025, 8 (01): : 233 - 240
  • [29] Representation Learning of Knowledge Graph with Semantic Vectors
    Gao, Tianyu
    Zhang, Yuanming
    Li, Mengni
    Lu, Jiawei
    Cheng, Zhenbo
    Xiao, Gang
    KNOWLEDGE SCIENCE, ENGINEERING AND MANAGEMENT, KSEM 2021, PT II, 2021, 12816 : 16 - 29
  • [30] Disentangled Relational Graph Neural Network with Contrastive Learning for knowledge graph completion
    Yin, Hong
    Zhong, Jiang
    Li, Rongzhen
    Li, Xue
    KNOWLEDGE-BASED SYSTEMS, 2024, 295