Learning from Context or Names? An Empirical Study on Neural Relation Extraction

被引:0
|
作者
Peng, Hao [1 ]
Gao, Tianyu [2 ]
Han, Xu [1 ]
Lin, Yankai [3 ]
Li, Peng [3 ]
Zhiyuan, Liu [1 ]
Sun, Maosong [1 ]
Zhou, Jie [3 ]
机构
[1] Tsinghua Univ, Dept Comp Sci & Technol, Beijing, Peoples R China
[2] Princeton Univ, Princeton, NJ USA
[3] Tencent Inc, Pattern Recognit Ctr, WeChat, Shenzhen, Peoples R China
基金
中国国家自然科学基金;
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Neural models have achieved remarkable success on relation extraction (RE) benchmarks. However, there is no clear understanding which type of information affects existing RE models to make decisions and how to further improve the performance of these models. To this end, we empirically study the effect of two main information sources in text: textual context and entity mentions (names). We find that (i) while context is the main source to support the predictions, RE models also heavily rely on the information from entity mentions, most of which is type information, and (ii) existing datasets may leak shallow heuristics via entity mentions and thus contribute to the high performance on RE benchmarks. Based on the analyses, we propose an entity-masked contrastive pre-training framework for RE to gain a deeper understanding on both textual context and type information while avoiding rote memorization of entities or use of superficial cues in mentions. We carry out extensive experiments to support our views, and show that our framework can improve the effectiveness and robustness of neural models in different RE scenarios. All the code and datasets are released at https://github.com/thunlp/RE-Context-or-Names.
引用
收藏
页码:3661 / 3672
页数:12
相关论文
共 50 条
  • [1] Bridging Research Fields: An Empirical Study on Joint, Neural Relation Extraction Techniques
    Ackermann, Lars
    Neuberger, Julian
    Kaeppel, Martin
    Jablonski, Stefan
    ADVANCED INFORMATION SYSTEMS ENGINEERING, CAISE 2023, 2023, 13901 : 471 - 486
  • [2] Context and Type Enhanced Representation Learning for Relation Extraction
    Yu, Erxin
    Jia, Yantao
    Wang, Shang
    Li, Fengfu
    Chang, Yi
    11TH IEEE INTERNATIONAL CONFERENCE ON KNOWLEDGE GRAPH (ICKG 2020), 2020, : 329 - 335
  • [3] LEARNING MENTION AND RELATION REPRESENTATION WITH CONVOLUTIONAL NEURAL NETWORKS FOR RELATION EXTRACTION
    Liang, Shuohong
    Chen, Guang
    Wang, Wei
    PROCEEDINGS OF 2016 5TH IEEE INTERNATIONAL CONFERENCE ON NETWORK INFRASTRUCTURE AND DIGITAL CONTENT (IEEE IC-NIDC 2016), 2016, : 437 - 441
  • [4] Improving Neural Relation Extraction with Positive and Unlabeled Learning
    He, Zhengqiu
    Chen, Wenliang
    Wang, Yuyi
    Zhang, Wei
    Wang, Guanchun
    Zhang, Min
    THIRTY-FOURTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THE THIRTY-SECOND INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE AND THE TENTH AAAI SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2020, 34 : 7927 - 7934
  • [5] Structured Minimally Supervised Learning for Neural Relation Extraction
    Bai, Fan
    Ritter, Alan
    2019 CONFERENCE OF THE NORTH AMERICAN CHAPTER OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS: HUMAN LANGUAGE TECHNOLOGIES (NAACL HLT 2019), VOL. 1, 2019, : 3057 - 3069
  • [6] A neural network framework for relation extraction: Learning entity semantic and relation pattern
    Zheng, Suncong
    Xu, Jiaming
    Zhou, Peng
    Bao, Hongyun
    Qi, Zhenyu
    Xu, Bo
    KNOWLEDGE-BASED SYSTEMS, 2016, 114 : 12 - 23
  • [7] Incorporating Human Knowledge in Neural Relation Extraction with Reinforcement Learning
    Liu, Bing
    Qi, Guilin
    Pan, Lu
    Duan, Shungfu
    Wu, Tianxing
    2019 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2019,
  • [8] Machine Learning and Deep Learning based Information Extraction from Bangla Names
    Kowsher, Md
    Sanjid, Md Zahidul Islam
    Das, Avishek
    Ahmed, Mahid
    Sarker, Md Murad Hossain
    9TH INTERNATIONAL YOUNG SCIENTISTS CONFERENCE IN COMPUTATIONAL SCIENCE, YSC2020, 2020, 178 : 224 - 233
  • [9] ARElight: Context Sampling of Large Texts for Deep Learning Relation Extraction
    Rusnachenko, Nicolay
    Liang, Huizhi
    Kalameyets, Maksim
    Shi, Lei
    ADVANCES IN INFORMATION RETRIEVAL, ECIR 2024, PT V, 2024, 14612 : 229 - 235
  • [10] RECON: Relation Extraction using Knowledge Graph Context in a Graph Neural Network
    Bastos, Anson
    Nadgeri, Abhishek
    Singh, Kuldeep
    Mulang, Isaiah Onando
    Shekarpour, Saeedeh
    Hoffart, Johannes
    Kaul, Manohar
    PROCEEDINGS OF THE WORLD WIDE WEB CONFERENCE 2021 (WWW 2021), 2021, : 1673 - 1685