Abstract
Relational triple extraction (RTE) from unstructured textual data is a pivotal operation in the field of information extraction. Among existing methods, tagging based extraction methods are gaining more and more attention. Nonetheless, the majority of these methods are plagued by the limitations of inadequate generalizability of span-based extraction techniques and the prediction of redundant relations. In addressing these challenges, we introduce a Redundant relation reduced bidirectional extraction framework based on SpanBERT for RTE (RSRTE). This framework uses SpanBERT as the encoder to obtain richer contextual information through span representation and better under-stand the complex relations between entities. During the decoding process, we construct a sentence relation prediction component, which extracts a subset of predicted relations of the input sentence to minimize the calculation of redundant relations. In addition, a bidirectional extraction method with a shared encoder component is used to extract all possible subject-object pairs from both directions, reducing the impact of incomplete subject extraction in one direction. Next, we use a biaffine based component to extract relations of entity pairs from the subset of predicted relations. Comprehensive experimental evaluations demonstrate that our proposed framework outperforms competitors on publicly available benchmark datasets and markedly enhances F1 scores in intricate situations involving overlapping triples and the extraction of multiple relations, thereby substantiating its efficacy.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Wei, Z., Su, J., Wang, Y., et al.: A novel cascade binary tagging framework for relational triple extraction. In: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (2020). https://doi.org/10.18653/v1/2020.acl-main.136
Ren, F., Zhang, L., Zhao, X., et al.: A simple but effective bidirectional framework for relational triple extraction. In: Proceedings of the Fifteenth ACM International Conference on Web Search and Data Mining, vol. 2022, pp. 824–832 (2022)
Zheng, H., Wen, R., Chen, X., et al.: PRGC: potential relation and global correspondence based joint relational triple extraction (2021). https://doi.org/10.18653/v1/2021.acl-long.486
Wang, Y., Yu, B., Zhang, Y., et al.: TPLinker: single-stage joint extraction of entities and relations through token pair linking (2020). https://doi.org/10.18653/v1/2020.coling-main.138
Yu, B., Zhang, Z., Shu, X., et al.: Joint extraction of entities and relations based on a novel decomposition strategy (2019). https://doi.org/10.48550/arXiv.1909.04273
Sun, K., Zhang, R., Mensah, S., et al.: Recurrent interaction network for jointly extracting entities and classifying relations. arXiv e-prints (2019). https://doi.org/10.18653/v1/2020.emnlp-main.304
Sun, K., Zhang, R., Mensah, S., et al.: Progressive multi-task learning with controlled information flow for joint entity and relation extraction. In: Proc. AAAI Conf. Artific. Intell. 35(15): 13851–13859 (2021)
Tian, X., Jing, L., He, L., et al.: Stereorel: relational triple extraction from a stereoscopic perspective. In: Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), vol. 2021, pp. 4851–4861 (2021)
Sui, D., Zeng, X., Chen, Y., et al.: Joint entity and relation extraction with set prediction networks. IEEE Trans. Neural Networks Learn. Syst. (2023)
Chen, Y., Zhang, Y., Hu, C., et al.: Jointly extracting explicit and implicit relational triples with reasoning pattern enhanced binary pointer network. In: North American Chapter of the Association for Computational Linguistics. Association for Computational Linguistics (2021). https://doi.org/10.18653/V1/2021.NAACL-MAIN.453
Shang, Y, M., Huang, H., Sun, X., et al.: Relational triple extraction: one step is enough. arXiv preprint arXiv:2205.05270 (2022)
Dai, Q., Yang, W., Wang, L., et al.: SOIRP: subject-Object Interaction and Reasoning Path based joint relational triple extraction by table filling. Neurocomputing 580, 127492 (2024)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Zhang, H., Xiang, J., Guo, X., Wang, L., Wang, X. (2024). A Redundant Relation Reduced Bidirectional Extraction Framework Based on SpanBERT for Relational Triple Extraction. In: Huang, DS., Chen, W., Zhang, Q. (eds) Advanced Intelligent Computing Technology and Applications. ICIC 2024. Lecture Notes in Computer Science, vol 14874. Springer, Singapore. https://doi.org/10.1007/978-981-97-5618-6_8
Download citation
DOI: https://doi.org/10.1007/978-981-97-5618-6_8
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-97-5617-9
Online ISBN: 978-981-97-5618-6
eBook Packages: Computer ScienceComputer Science (R0)