ABSTRACT
Entity relation extraction aims to extract knowledge triples from unstructured or semi-structured text data and can be applied to various fields, including medicine, finance knowledge graph construction and intelligent question-answering. Traditional entity relation extraction requires a large amount of labeled data, consumes a lot of labor and time, and the trained model lacks generalization ability, which is difficult to migrate to other fields. Zero-shot entity relation extraction relieves the dependence on labeled data in traditional method. Based on unlabeled text data, zero-shot entity relation extraction has strong domain adaptability, which is a very challenging and practical task. Recent work on large language models shows that large models can effectively complete downstream tasks through natural language instructions and have good generalization ability. Inspired by this, we explore the use of large models for information extraction. Due to the randomness of large language model generation, we introduce in-context learning in entity relation extraction task to guide large language model to output data in a specified format to help obtain structured data. At the same time, we propose a three-stage extraction framework for decomposing entity relation extraction tasks, and each stage is conducted in the form of question and answer to reduce the complexity of extraction. We evaluated the knowledge triples extraction performance of the model on three self-built test datasets in different fields, and the experimental result showed that our proposed method achieved impressive performance in the zero-shot entity relation extraction task, surpassing the comparison model on multiple metrics, proving the effectiveness and domain adaptability of the proposed method.
- Brown, T., Mann, B., Ryder, N., Subbiah, M., Kaplan, J. D., Dhariwal, P., ... & Amodei, D. 2020. Language models are few-shot learners. Advances in neural information processing systems, 33, 1877-1901.Google Scholar
- Ouyang, L., Wu, J., Jiang, X., Almeida, D., Wainwright, C., Mishkin, P., ... & Lowe, R. 2022. Training language models to follow instructions with human feedback. Advances in Neural Information Processing Systems, 35, 27730-27744.Google Scholar
- Chinchor, N., & Marsh, E. 1998, July. Muc-7 information extraction task definition. In Proceeding of the seventh message understanding conference (MUC-7), Appendices (pp. 359-367).Google Scholar
- Hinton, G. E., & Salakhutdinov, R. R. (2006). Reducing the dimensionality of data with neural networks. science, 313(5786), 504-507.Google Scholar
- Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., ... & Polosukhin, I. 2017. Attention is all you need. Advances in neural information processing systems, 30.Google Scholar
- Devlin, J., Chang, M. W., Lee, K., & Toutanova, K. 2018. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805.Google Scholar
- Sun, Y., Wang, S., Li, Y., Feng, S., Chen, X., Zhang, H., ... & Wu, H. 2019. Ernie: Enhanced representation through knowledge integration. arXiv preprint arXiv:1904.09223.Google Scholar
- Zhong, Z., & Chen, D. 2020. A frustratingly easy approach for entity and relation extraction. arXiv preprint arXiv:2010.12812.Google Scholar
- Wei, Z., Su, J., Wang, Y., Tian, Y., & Chang, Y. 2019. A novel cascade binary tagging framework for relational triple extraction. arXiv preprint arXiv:1909.03227.Google Scholar
- Wang, Y., Yu, B., Zhang, Y., Liu, T., Zhu, H., & Sun, L. 2020. TPLinker: Single-stage joint extraction of entities and relations through token pair linking. arXiv preprint arXiv:2010.13415.Google Scholar
- Zeng, X., Zeng, D., He, S., Liu, K., & Zhao, J. 2018, July. Extracting relational facts by an end-to-end neural model with copy mechanism. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (pp. 506-514).Google ScholarCross Ref
- Zeng, D., Liu, K., Chen, Y., & Zhao, J. 2015, September. Distant supervision for relation extraction via piecewise convolutional neural networks. In Proceedings of the 2015 conference on empirical methods in natural language processing (pp. 1753-1762).Google ScholarCross Ref
- Lu, Y., Liu, Q., Dai, D., Xiao, X., Lin, H., Han, X., ... & Wu, H. 2022. Unified structure generation for universal information extraction. arXiv preprint arXiv: 2203.12277.Google Scholar
- Wei, X., Cui, X., Cheng, N., Wang, X., Zhang, X., Huang, S., ... & Han, W. 2023. Zero-shot information extraction via chatting with chatgpt. arXiv preprint arXiv: 2302.10205.Google Scholar
- Zeng, A., Liu, X., Du, Z., Wang, Z., Lai, H., Ding, M., ... & Tang, J. 2022. Glm-130b: An open bilingual pre-trained model. arXiv preprint arXiv:2210.02414.Google Scholar
Index Terms
- LlmRe: A zero-shot entity relation extraction method based on the large language model
Recommendations
An Exploration of Prompt-Based Zero-Shot Relation Extraction Method
Chinese Computational LinguisticsAbstractZero-shot relation extraction is an important method for dealing with the newly emerging relations in the real world which lacks labeled data. However, the mainstream two-tower zero-shot methods usually rely on large-scale and in-domain labeled ...
A Few-Shot Relation Extraction Method for Enhancing Entity Attention
CSSE '22: Proceedings of the 5th International Conference on Computer Science and Software EngineeringThe aim of the few-shot relation extraction (FSRE) method is to study the relation classification problem with fewer samples. An effective few-shot relation extraction model EnAttConceptFERE is proposed to effectively classify relationships through ...
A Relation-Oriented Method for Joint Entity and Relation Extraction Based on Neural Network
EITCE '21: Proceedings of the 2021 5th International Conference on Electronic Information Technology and Computer EngineeringEntity and relation extraction is a basic task of information extraction in natural language processing. At present, Entity and relation extraction based on artificial intelligence has been widely studied, but most methods adopt the idea of identifying ...
Comments