基于审判逻辑步骤的裁判文书摘要生成方法

doi:10.3778/j.issn.1002-8331.2209-0142

摘要/Abstract

摘要： 面向裁判文书的司法摘要是提升裁判文书分析能力的关键技术。裁判文书作为审判活动的载体，精准地呈现了案件的审判逻辑，但目前针对裁判文书的摘要方法只关注裁判文书的序列化信息，忽视了裁判文书的逻辑结构，且不能有效解决文本过长、信息冗余等问题。提出基于审判逻辑步骤的裁判文书摘要生成方法，采取“抽取+生成”相结合的方式，在抽取部分利用多标签分类方法，依据人民法院审理案件的逻辑步骤抽取出“类型、诉请、事实、结果”四个句子集合，在生成部分由微调后的T5-PEGASUS模型得到摘要。利用基于内部知识的最大相似度匹配算法对“事实”部分的输入文本进行降噪处理，进一步改善了摘要效果。实验结果表明，相比于主流的指针生成网络模型，该方法在ROUGE-1、ROUGE-2和ROUGE-L的F1指标上分别提升了17.99个百分点、21.24个百分点、21.86个百分点，说明在司法摘要任务中引入逻辑结构能够提升性能。

关键词: 裁判文书, 审判逻辑步骤, 多标签分类, 内部知识, 生成式摘要

Abstract: Judicial summary oriented to judgment documents is the key technology to improve the analytical ability of judgment documents. As the carrier of the trial activities, the judgment documents accurately present the trial logic of the case. However, the current abstract methods only focus on the serialization information of the judgment documents, ignore the logical structure, and can not effectively solve the problems of too long texts and redundant information. A judgment document summary generation method based on the trial logic steps is proposed. The method of “extraction + generation” is adopted. The extraction part uses the multi-label classification method to extract four sentence sets of “type, claim, fact and result” according to the logic steps of the people's court. The generation part gets the summary from the fine-tuned T5-PEGASUS model. And the input text of the “fact” part is denoised by using the maximum similarity matching algorithm based on internal knowledge, which further improves the summary effect. The experimental results show that, compared with the mainstream pointer-generated network summary model, the proposed method improves the F1 index of ROUGE-1, ROUGE-2 and ROUGE-L by 17.99 percentage points, 21.24 percentage points and 21.86 percentage points, respectively. This shows that introducing logical structure into the judicial summarization can improve the performance of the task.

Key words: judgment document, trial logic steps, multi-label classification, internal knowledge, abstractive summary

余帅, 宋玉梅, 秦永彬, 黄瑞章, 陈艳平. 基于审判逻辑步骤的裁判文书摘要生成方法[J]. 计算机工程与应用, 2024, 60(4): 113-121.

YU Shuai, SONG Yumei, QIN Yongbin, HUANG Ruizhang, CHEN Yanping. Method for Generating Summary of Judgment Documents Based on Trial Logic Steps[J]. Computer Engineering and Applications, 2024, 60(4): 113-121.

参考文献

[1] Supreme People's Court. Provisions of the Supreme People??s Court on the publication of judgments on the Internet by the People??s Courts (2016 revision) [EB/OL]. (2016-10-01)[2022-07-02]. https://pkulaw.com/.
[2] WIDYASSARI A P, RUSTAD S, SHIDIK G F, et al. Review of automatic text summarization techniques & methods[J]. Journal of King Saud University (Computer and Information Sciences) 2022, 34(4): 1029-1046.
[3] MORATANCH N, CHITRAKALA S. A survey on extractive text summarization[C]//Proceedings of the 2017 International Conference on Computer, Communication and Signal Processing, 2017: 1-6.
[4] AKHIL R. Extractive text summarization by deep learning[J]. arXiv: 1708. 04439, 2017.
[5] ZHANG Y, LI D, WANG Y H, et al. Abstract text summarization with a convolutional seq2seq model[J]. Applied Sciences, 2019, 9(8): 1665.
[6] LUO D, XING C P. Logical formula of trial in civil and commercial cases[N/OL]. People??s court newspaper (2018-04-04)[2022-07-02]. http://rmfyb.chinacourt.org/paper/html/2018-04/04/content_137571. htm.
[7] HELLI B, Moghaddam M E. A text-independent Persian writer identification system using LCS based classifier[C]//Proceedings of the 2008 IEEE International Symposium on Signal Processing and Information Technology, 2008: 203-206.
[8] LUHN H P. The automatic creation of literature abstracts[J]. IBM Journal of Research and Development, 1958, 2(2): 159-165.
[9] MIHALCEA R, TARAU P. TextRank: bringing order into text[C]//Proceedings of the 2004 Conference on Empirical Methods in Natural Language Processing, 2004: 404-411.
[10] PAGE L. The PageRank citation ranking: bringing order to the Web[R]. Stanford Digital Library Technologies Project, 1998.
[11] XIAO W, CARENINI G. Extractive summarization of long documents by combining global and local context[J]. arXiv: 1909. 08089, 2019.
[12] KIM Y, RUSH A M. Sequence-level knowledge distillation[J]. arXiv:1606. 07947, 2016.
[13] KIM M, SINGH M D, LEE M. Towards abstraction from extraction: multiple timescale gated recurrent unit for summarization[J]. arXiv:1607. 00718, 2016.
[14] SEE A, LIU P J, MANNING C D. Get to the point: summarization with pointer-generator networks[J]. arXiv:1704. 04368, 2017.
[15] FARZINDAR A, LAPALME G. LetSum, an automatic legal text summarizing[C]//Proceedings of the 17th Annual Conference on Legal Knowledge and Information Systems. Amsterdam: IOS Press, 2004: 11-18.
[16] HACHEY B, GROVER C. Extractive summarisation of legal texts[J]. Artificial Intelligence and Law, 2006, 14(4): 305-345.
[17] POLSLEY S, JHUNJHUNWALA P, HUANG R. CaseSummarizer: a system for automated summarization of legal texts[C]//Proceedings of the 26th International Conference on Computational Linguistics: System Demonstrations, 2016: 258-262.
[18] REN P J, CHEN Z M, REN Z C, et al. Leveraging contextual sentence relations for extractive summarization using a neural attention model[C]//Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval, 2017: 95-104.
[19] 张迎, 王中卿, 王红玲. 基于篇章主次关系的单文档抽取式摘要方法研究[J]. 中文信息学报, 2019, 33(8): 67-76.
ZHANG Y, WANG Z Q, WANG H L. Single document extractive summarization with satellite and nuclear relations[J]. Journal of Chinese Information Processing, 2019, 33(8): 67-76.
[20] WU R S, ZHANG Y F, WANG H L, et al. Abstractive summarization based on hierarchical structure[J]. Journal of Chinese Information Processing, 2019, 33(8): 90-98.
[21] LI Y Y, KRISHNAMURTHY R, RAGHAVAN S, et al. Regular expression learning for information extraction[C]//Proceedings of the 2008 Conference on Empirical Methods in Natural Language Processing, 2008: 21-30.
[22] CUI L Y, ZHANG Y. Hierarchically-refined label attention network for sequence labeling[J]. arXiv:1908.08676, 2019.
[23] TSOUMAKAS G, KATAKIS I. Multi-label classification: an overview[J]. International Journal of Data Warehousing and Mining, 2007, 3(3): 1-13.
[24] RAFFEL C, SHAZEER N, ROBERTS A, et al. Exploring the limits of transfer learning with a unified text-to-text transformer[J]. The Journal of Machine Learning Research, 2020, 21: 5485-5551.
[25] ZHANG J Q, ZHAO Y, SALEH M, et al. PEGASUS: pre-training with extracted gap-sentences for abstractive summarization[C]//Proceedings of the 37th International Conference on Machine Learning, 2020: 11328-11339.
[26] DEVLIN J, CHANG M W, LEE K, et al. BERT: pre-training of deep bidirectional transformers for language understanding[J]. arXiv: 1810. 04805, 2018.
[27] LIN M, CHEN Q, YAN S. Network in network[J]. arXiv: 1312. 4400, 2013.
[28] ZHANG J J, FANG M, LI X. Multi-label learning with discriminative features for each label[J]. Neurocomputing, 2015, 154: 305-316.
[29] LIN C Y, HOVY E. Automatic evaluation of summaries using n-gram co-occurrence statistics[C]//Proceedings of the 2003 Human Language Technology Conference of the North American Chapter of the Association for Computational Linguistics, 2003: 150-157.
[30] LEWIS M, LIU Y, GOYAL N, et al. BART: denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020: 7871-7880.