Joint Information Extraction Model Based on Feature Sharing

Zhangchi Gao; Shoubin Li

doi:10.62677/IJETAA.2402107

Authors

Zhangchi Gao Institute of Software, Chinese Academy of Sciences Author
Shoubin Li Institute of Software, Chinese Academy of Sciences Author

DOI:

https://doi.org/10.62677/IJETAA.2402107

Keywords:

Natural Language Processing, Information Extraction, Entity Extraction, Relation Extraction, Event Extraction

Abstract

To address the challenge of efficiently and accurately extracting entities, relationships, and events from unstructured text, a joint information extraction model based on feature sharing is proposed. This model utilizes the contextual information of entities, relationships, and events, and integrates entity extraction, relationship extraction, and event extraction tasks through a multi-feature cascade encoder to achieve joint extraction. To validate the effectiveness of the model, comparative analysis was conducted on military news datasets, comparing against two typical information extraction models. Results demonstrated superiority over current state-of-the-art baselines.

Downloads

Download data is not yet available.

References

Junyi Wang et al. "Method and device for portfolio analysis based on knowledge graph." Patent CN202211181071.7, 2023.

Eddie Chen and Xi Cai. "Construction of knowledge map for large-scale construction project management." Journal of Changsha University of Technology (Natural Science Edition), 2016.

Xiaoyan Zhang, Ting Wang, and Huowang Chen. "A study of named entity recognition." Computer Science 32.4 (2005): 5.

Hongkui Yu et al. "Chinese named entity recognition based on cascading hidden Markov models." Journal of Communications 27.2 (2006): 8.

Yuntian Feng, Hongjun Zhang, Wenning Hao. "Named entity recognition for military text." Computer Science,2015, 42(7): 15-18.

Jing Dong et al. "A study of feature selection in Chinese entity-relationship extraction." Journal of Chinese Information Processing 21.4 (2007): 80-91.

Fengshuai Gao and Huabin Yang. "A military named entity relationship extraction method combining word rules and SVM modeling." Information Communications 11 (2017): 2.

Wenyu Duan et al. "Research on entity and relationship extraction methods for weapon and equipment domain." Journal of the Chinese Academy of Electronics and Information Science 17.12 (2022): 1165-1172.

Yanyan Zhao et al. "Research on Chinese event extraction techniques." Journal of Chinese Information Processing 22.1 (2008): 3-8.

Yuanfang Yu et al. "Multi-round event argument extraction guided by role information." Journal of Peking University (Natural Science) 59.1 (2023): 83-91.

Qi Li et al. "Constructing information networks using one single model." Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2014.

Bishan Yang and Tom Mitchell. "Joint extraction of events and entities within a document context." 2016.

Wentao Wu, Peifeng Li, and Qiaoming Zhu. "A joint entity and event extraction method based on hybrid neural networks." Journal of Chinese Information Processing 33.8 (2019): 7.

Jacob Devlin et al. "BERT: pre-training of deep bidirectional transformers for language understanding." 2018.