Page 295 - 《软件学报》2025年第10期

4692 软件学报 2025 年第 36 卷第 10 期

自动化规约提供了一个有效途径, 也为其他领域知识密集系统的知识嵌入提供了有价值的借鉴.
在未来的工作中, 我们计划不断提高抽取模型的准确率, 丰富领域知识库, 由此进一步提高自动化生成的需求
规约质量, 能更快地应对业务规则的变化. 特别地, 由于证券行业的要求, 未来也考虑建立领域化的大语言模型, 以
便部署到证券公司中, 真正投入使用. 同时, 我们考虑将该方法拓展到证券领域以外的其他行业中去.

References:
[1] Shenzhen Stock Exchange. Shenzhen Stock Exchange Bond trading rules. 2022 (in Chinese). https://docs.static.szse.cn/www/lawrules/
rule/bond/bonds/trade/W020220127631859244722.pdf
[2] Shanghai Stock Exchange. Trading rules of Shanghai Stock Exchange (Revised in 2023). 2023 (in Chinese). http://www.sse.com.cn/
lawandrules/sselawsrules/trade/universal/c/10118241/files/e773c99a2fef4fbc930bc332492156d8.docx
[3] Sleimi A, Sannier N, Sabetzadeh M, Briand L, Ceci M, Dann J. An automated framework for the extraction of semantic legal metadata
from legal texts. Empirical Software Engineering, 2021, 26(3): 43. [doi: 10.1007/s10664-020-09933-5]
[4] Sannier N, Adedjouma M, Sabetzadeh M, Briand L. An automated framework for detection and resolution of cross references in legal
texts. Requirements Engineering, 2017, 22(2): 215–237. [doi: 10.1007/s00766-015-0241-3]
[5] Bhuiyan H, Governatori G, Bond A, Rakotonirainy A. Traffic rules compliance checking of automated vehicle maneuvers. Artificial
Intelligence and Law, 2024, 32(1): 1–56. [doi: 10.1007/s10506-022-09340-9]
[6] Chen XH, Yin B, Jin Z. Ontology-guided requirements modeling based on problem frames approach. Ruan Jian Xue Bao/Journal of
Software, 2011, 22(2): 177–194 (in Chinese with English abstract). http://www.jos.org.cn/1000-9825/3755.htm [doi: 10.3724/SP.J.1001.
2011.03755]
[7] Chen XH, Jin Z. An ontology-guided process for developing problem frame specification: An example. In: Proc. of the 3rd Int’l
Workshop on Applications and Advances of Problem Frames. Leipzig: ACM, 2008. 36–39. [doi: 10.1145/1370811.1370818]
[8] Bencharqui H, Haidrar S, Anwar A. Ontology-based requirements specification process. In: Proc. of the 10th Int’l Conf. on Innovation,
Modern Applied Science & Environmental Studies. 2022. 01045. [doi: 10.1051/e3sconf/202235101045]
[9] Abdalazeim A, Meziane F. A review of the generation of requirements specification in natural language using objects UML models and
domain ontology. Procedia Computer Science, 2021, 189: 328–334. [doi: 10.1016/j.procs.2021.05.102]
[10] Zhang ZH, Fang M, Chen L, Namazi-Red MR, Wang J. How do large language models capture the ever-changing world knowledge? A
review of recent advances. In: Proc. of the 2023 Conf. on Empirical Methods in Natural Language Processing. Singapore: ACL, 2023.
8289–8311. [doi: 10.18653/v1/2023.emnlp-main.516]
[11] Kau A, He XZ, Nambissan A, Astudillo A, Yin H, Aryani A. Combining knowledge graphs and large language models. arXiv:2407.
06564, 2024.
[12] Lu RS, Lin CC, Tsao HY. Empowering large language models to leverage domain-specific knowledge in E-learning. Applied Sciences,
2024, 14(12): 5264. [doi: 10.3390/app14125264]
[13] Pan SR, Luo LH, Wang YF, Chen C, Wang JP, Wu XD. Unifying large language models and knowledge graphs: A roadmap. IEEE Trans.
on Knowledge and Data Engineering, 2024, 36(7): 3580–3599. [doi: 10.1109/TKDE.2024.3352100]
[14] OpenAI. ChatGPT. 2024. https://chatgpt.com
[15] Knauf R, Spreeuwenberg S, Gerrits R, Jendreck M. A step out of the ivory tower: Experiences with adapting a test case generation idea to
business rules. In: Proc. of the 17th Int’l Florida Artificial Intelligence Research Society Conf. Miami Beach: AAAI, 2004. 343–348.
[16] Li XZ, Chan S, Zhu XD, Pei YL, Ma ZQ, Liu XM, Shah S. Are ChatGPT and GPT-4 general-purpose solvers for financial text analytics?
A study on several typical tasks. In: Proc. of the 2023 Conf. on Empirical Methods in Natural Language Processing: Industry Track.
Singapore: ACL, 2023. 408–422. [doi: 10.18653/v1/2023.emnlp-industry.39]
[17] Wei J, Zou K. EDA: Easy data augmentation techniques for boosting performance on text classification tasks. In: Proc. of the 2019 Conf.
on Empirical Methods in Natural Language Processing and the 9th Int’l Joint Conf. on Natural Language Processing. Hong Kong: ACL,
2019. 6382–6388. [doi: 10.18653/v1/D19-1670]
[18] Zhang ZS, Zhang HQ, Chen KM, Guo YH, Hua JY, Wang YL, Zhou M. Mengzi: Towards lightweight yet ingenious pre-trained models
for Chinese. arXiv:2110.06696, 2021.
[19] Loshchilov I, Hutter F. Decoupled weight decay regularization. In: Proc. of the 7th Int’l Conf. on Learning Representations. 2019.
[20] Liu L, Özsu MT. Encyclopedia of Database Systems. 2nd ed., New York: Springer, 2018. 2245–2248. [doi: 10.1007/978-1-4614-8265-9]
[21] He H, Choi JD. The stem cell hypothesis: Dilemma behind multi-task learning with transformer encoders. In: Proc. of the 2021 Conf. on
Empirical Methods in Natural Language Processing. Punta Cana: ACL, 2021. 5555–5577. [doi: 10.18653/v1/2021.emnlp-main.451]
[22] Shenzhen Stock Exchange. Special provisions for trading on the ChiNext Board of Shenzhen Stock Exchange. 2023 (in Chinese). https://
www.szse.cn/lawrules/rule/repeal/rules/P020231230545310237980.pdf