基于有序多任务学习的司法二审判决预测方法

PDF(2910 KB)

中文信息学报 ›› 2022, Vol. 36 ›› Issue (3) : 162-172.

情感分析与社会计算

基于有序多任务学习的司法二审判决预测方法

韩晓晖^1,2,王文同³,宋连欣¹,刘广起^1,2,崔超然⁴,尹义龙³

作者信息 +

Second Instance Judgement Prediction Based on Sequential Multi-task Learning

HAN Xiaohui^1,2, WANG Wentong³, SONG Lianxin¹, LIU Guangqi^1,2, CUI Chaoran⁴, YIN Yilong³

Author information +

History +

摘要

司法二审判决预测任务旨在基于一审判决、新发现事实、上诉理由等文本材料预测二审程序的判决结果,其难点在于如何捕捉两审法院对案件事实的认知异同来生成可解释的预测。针对上述难点,该文提出一种基于有序多任务学习的二审判决预测方法SIJP-SML,该方法通过两个时序依赖的多任务学习部分对一审到二审的完整审判逻辑进行建模,以提取并融合一、二审法院对案件事实的认知表示来预测二审判决。同时,SIJP-SML在多任务学习中引入法院观点生成任务来输出具有一定可读性的判决理据,以增强预测的可解释性。在6万余份二审裁判文书数据上的实验结果证明了SIJP-SML的有效性和合理性,其综合性能优于所有基线方法。

Abstract

The second instance judgment prediction task aims to predict the judgment results of appeal trials based on the first instance judgment, newly discovered facts, and appeal reasons. There are two challenges to solve the second instance judgment prediction task. One is how to capture the cognitive similarities and differences between superior and lower courts on case facts. The other is how to make the prediction interpretable. To address these challenges, we propose SIJP-SML, which is a second instance decision prediction framework based on sequential multi-task learning. SIJP-SML models the complete trial logic from the first instance to the second instance through two time-dependent multi-task learning components. Cognitive representations of superior and lower courts are extracted from case facts and integrated to produce the second trial decision. To improve prediction interpretability, SIJP-SML takes court-view-generation as one of the learning tasks to output judgment rationales with readability. Experimental results on a dataset of more than 60K second instance judgment documents reveal that SIJP-SML outperforms all the baseline methods.

导出引用

韩晓晖,王文同,宋连欣,刘广起,崔超然,尹义龙. 基于有序多任务学习的司法二审判决预测方法. 中文信息学报. 2022, 36(3): 162-172

HAN Xiaohui, WANG Wentong, SONG Lianxin, LIU Guangqi, CUI Chaoran, YIN Yilong. Second Instance Judgement Prediction Based on Sequential Multi-task Learning. Journal of Chinese Information Processing. 2022, 36(3): 162-172

参考文献

[1] Long S,Tu C,Liu Z,et al. Automatic judgment prediction via legal reading comprehension[C]//Proceedings of China National Conference on Chinese Computational Linguistics,2019: 558-572.
[2] Wang P,Yang Z,Niu S,et al. Modeling dynamic pairwise attention for crime classification over legal articles[C]//Proceedings of the 41st International ACM SIGIR Conference on Research and Development in Information Retrieval, 2018: 485-494.
[3] Wang P,Fan Y,Niu S,et al. Hierarchical matching network for crime classification[C]//Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval, 2019: 325-334.
[4] Luo B,Feng Y,Xu J,et al. Learning to predict charges for criminal cases with legal basis[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing,2017: 2727-2736.
[5] Hu Z,Li X,Tu C,et al. Few-shot charge prediction with discriminative legal attributes[C]//Proceedings of the 27th International Conference on Computational Linguistics,2018: 487-498.
[6] 谭红叶,张博文,张虎,等.面向法律文书的量刑预测方法研究[J].中文信息学报,2020,2034(003): 107-114.
[7] Chen H,Cai D,Dai W,et al. Charge-based prison term prediction with deep gating network[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing,2019: 6361-6366.
[8] 马康.反腐败视野下刑事二审程序纠错功能研究[J].山东法官培训学院学报,2020,36(03): 143-152.
[9] Katz D M,Bommarito M J,Blackman J. A general approach for predicting the behavior of the supreme court of the united states[J]. PloS ONE,2017,12(4): e0174698: 1-18.
[10] Ye H,Jiang X,Luo Z,et al. Interpretable charge predictions for criminal cases: learning to generate court views from fact descriptions[C]//Proceedings of the Conference of the North American Chapter of the Association for Computational Linguistics,2018: 1854-1864.
[11] Kort F. Predicting supreme court decisions mathematically: a quantitative analysis of the “right to counsel” cases[J]. The american political science review,1957,51(1): 1-12.
[12] Ulmer S S. Quantitative analysis of judicial processes: some practical and theoretical applications[J]. Law and contemporary problems,1963,28(1): 164-184.
[13] R Keown. Mathematical models for legal prediction[J]. Computer Law Journal,1980,2(1): 829-862.
[14] Segal J A. Predicting supreme court cases probabilistically: the search and seizure cases(1962-1981)[J]. American political science review,1984,78(4): 891-900.
[15] Lauderdale B E,Clark T S. The supreme courts many median justices[J]. American political science review,2012,106(4): 847-866.
[16] Shapira M. Computerized decision technology in social service[J]. International journal of sociology and social policy,1990,10(4/5/6): 138-164.
[17] Hassett P. Can expert system technology contribute to improved bail conditions?[J]. International journal of law and information technology,1993,1(2): 144-188.
[18] Lin W C,Kuo T T,Chang T J,et al. Exploiting machine learning models for Chinese legal documents labeling, case classification,and sentencing prediction [J]. International journal of computational linguistics & Chinese language processing,2012,17(4): 49-68.
[19] Aletras N,Tsarapatsanis D,Pretiuc Pietro D,et al. Predicting judicial decisions of the European court of human rights: a natural language processing perspective[J]. PeerJ Computer Science,2016,2(10): e9301-e9319.
[20] Sulea O M,Zampieri M,Malmasi S,et al. Exploring the use of text classification in the legal domain[C]//Proceedings of the 2nd Workshop on Automated Semantic Analysis of Information in Legal Texts, 2017.
[21] 刘宗林,张梅山,甄冉冉,等.融入罪名关键词的法律判决预测多任务学习模型[J].清华大学学报(自然科学版),2019,59(07): 497-504.
[22] Li S, Zhang H,Ye L,et al. MANN: a multichannel attentive neural network for legal judgment prediction[J]. IEEE Access,2019,7(1): 151144-151155.
[23] Xu N,Wang P,Chen L,et al. Distinguish confusing law articles for legal judgment prediction [C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics,2020: 3086-3095.
[24] Zhong H,Guo Z,Tu C,et al. Legal judgment prediction via topological learning [C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing,2018: 3540-3549.
[25] Yang W,Jia W,Zhou X I,et al. Legal judgment prediction via multi-perspective bi-feedback network [C]//Proceedings of the 28th International Joint Conference on Artificial Intelligence,2019: 4085-4091.
[26] Jiang X,Ye H,Luo Z,et al. Interpretable rationale augmented charge prediction system [C]//Proceedings of the 27th International Conference on Computational Linguistics: System Demonstrations. Association for Computational Linguistics,2018: 146-151.
[27] Yang X,Shi G,Lou J,et al. Interpretable charge prediction with multi-perspective jointly learning model [C]//Proceedings of the IEEE 5th International Conference on Computer and Communications,2019: 1850-1855.
[28] Chalkidis I,Androutsopoulos I,Aletras N. Neural legal judgment prediction in English [C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics,2019: 4317-4323.
[29] 杨杰辉. 刑事上诉理由研究[J]. 法治研究,2017(06): 128-135.
[30] Kingma D,Ba J. Adam: a method for stochastic optimization[C]//Proceedings of the 3rd International Conference for Learning Representations,2015.
[31] Kim Y. Convolutional neural networks for sentence classification [C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing,2014: 1746-1751.
[32] Sutskever I,Vinyals O,Le V Q. Sequence to sequence learning with neural networks [C]//Proceedings of the 27th International Conference on Neural Information Processing Systems,2014: 3104-3112.
[33] Yang Z,Yang D,Dyer C,et al. Hierarchical attention networks for document classification [C]//Proceedings of the Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies,2016: 1480-1489.

基金

国家重点研发计划(2018YFC0830100,2018YFC0830102);国家自然科学基金(61602281);山东省重点研发计划(2019JZZY010132)

PDF(2910 KB)

1546

Accesses

Citation

Detail

段落导航

摘要
Abstract
关键词
Key words
引用本文
参考文献
基金

Received	Published
2021-01-18	2022-04-20
Issue Date
2022-04-20

选择文件类型/文献管理软件名称

选择包含的内容

摘要

Abstract

关键词

Key words

引用本文

{{custom_sec.title}}

{{custom_sec.title}}

参考文献

{{custom_fnGroup.title_cn}}

脚注

基金