基于多特征融合的中医药问题生成模型
A Traditional Chinese Medicine Question Generation Model Based on Multi-Feature Fusion
摘要: 目的：提出一种基于多特征融合的中医药问题生成模型(MFFQG)，以改善现有的自动生成技术在处理特定领域时存在的领域关键词信息缺失和生成问题表达不规范问题。方法：利用RoBERTa向量和五笔向量捕捉输入序列的语义特征和字形特征，同时融合句法信息和所构建的中医药领域主副关键词信息，将得到的多特征向量信息送入UniLM生成模型得到生成结果，实现对中医药领域问题的自动生成。结果：MFFQG模型融合多种特征，在Rouge-1、Rouge-2、Rouge-L评价指标上分别达到64.93%、34.57%、63.05%。局限：数据主要来源于中医药领域，在其他领域中的效果有待验证。结论：MFFQG模型相较于对比模型，可以显著提升中医药问题的生成质量。
Abstract: Objective: To propose a traditional Chinese medicine problem generation model (MFFQG) based on multi feature fusion, in order to improve the problems of missing domain keyword information and non-standard expression of generation problems in existing automatic generation technologies when dealing with specific fields. Method: Using RoBERTa vectors and Wubi vectors to capture the semantic and glyph features of the input sequence, while integrating syntactic information and the constructed main and auxiliary keyword information in the field of traditional Chinese medicine, the obtained multi feature vector information is fed into the UniLM generation model to obtain the generated results, achieving automatic generation of problems in the field of traditional Chinese medicine. Result: The MFFQG model integrates multiple features and achieves 64.93%, 34.57%, and 63.05% in Rouge-1, Rouge-2, and Rouge-L evaluation indicators, respectively. Limitation: The data mainly comes from the field of traditional Chinese medicine, and its effectiveness in other fields needs to be verified. Conclusion: Compared to the comparative model, the MFFQG model can significantly improve the quality of generating traditional Chinese medicine problems.
文章引用：酒敬天, 李红莲. 基于多特征融合的中医药问题生成模型[J]. 人工智能与机器人研究, 2024, 13(3): 673-683. https://doi.org/10.12677/airr.2024.133068

参考文献

[1] Dhole, K. and Manning, C.D. (2020) Syn-QG: Syntactic and Shallow Semantic Rules for Question Generation. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, Online, 5-10 July 2020, 752-765. [Google Scholar] [CrossRef] 
[2] Labutov, I., Basu, S. and Vanderwende, L. (2015) Deep Questions without Deep Understanding. Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), Beijing, 26-31 July 2015, 889-898. [Google Scholar] [CrossRef] 
[3] Mazidi, K. and Nielsen, R.D. (2014) Linguistic Considerations in Automatic Question Generation. Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), Baltimore, 23-25 June 2014, 321-326. [Google Scholar] [CrossRef] 
[4] Heilman, M. and Smith, N.A. (2010) Good Question! Statistical Ranking for Question Generation. Proceedings of Human Language Technologies: The Annual Conference of the North American Chapter of the Association for Computational Linguistics, Los Angeles, 2-4 June 2010, 609-617.
[5] Chali, Y. and Hasana, A. (2012) Towards Automatic Topical Question Generation[C]//Proceedings of the 24th International Conference on Computational Linguistics, Mumbai, 2012: 475-492.
[6] Duan, J.-Y., Xu, L.-S., Liu, J., et al. (2022) Question Generation Based on Sememe Knowledge and Bidirectional Attention Flow. Data Analysis and Knowledge Discovery, 6, 44-53.
[7] Zeng, H., Zhi, Z., Liu, J. and Wei, B. (2021) Improving Paragraph-Level Question Generation with Extended Answer Network and Uncertainty-Aware Beam Search. Information Sciences, 571, 50-64. [Google Scholar] [CrossRef] 
[8] Hu, Y. and Zhou, G.-Y. (2022) Question Generation from Knowledge Base with Graph Transformer. Journal of Chinese Information Processing, 36, 111-120.
[9] Fei, Z., Zhang, Q. and Zhou, Y. (2021) Iterative GNN-Based Decoder for Question Generation. Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, Punta Cana, 7-11 November 2021, 2573-2582. [Google Scholar] [CrossRef] 
[10] Serban, I.V., Garcia-Durán, A., Gulcehre, C., et al. (2018) Generating Factoid Questions with Recurrent Neural Networks: The 30M Factoid Question-Answer Corpus. Proceedings of Annual Meeting of the Association for Computational Linguistics, Berlin, 7-12 August 2016, 588-598.
[11] Ma, H., Wang, J., Lin, H. and Xu, B. (2022) Graph Augmented Sequence-to-Sequence Model for Neural Question Generation. Applied Intelligence, 53, 14628-14644. [Google Scholar] [CrossRef] 
[12] Li, Y.-F., Ye, D.-Y. and Chen, Y.-Z. (2023) Knowledge Enhanced Biograph Interaction Neural Network for Question Generation. Journal of Chinese Computer Systems.

为你推荐