Review of Legal Judgment Prediction Based on Graph Deep Learning

doi:10.3778/j.issn.1673-9418.2409085

Abstract

Abstract: Legal judgment prediction aims to predict judgment results based on the description of case facts, such as law articles, charges, and term of penalty involved in the case. In recent years, graph deep learning methods have achieved remarkable success in the field of legal artificial intelligence, capturing complex relationships between case facts and the dependencies between law articles effectively. This paper summarizes the latest achievements in graph deep learning methods related to legal judgment prediction comprehensively. Firstly, this paper introduces the task definition and early research of legal judgment prediction. Secondly, this paper focuses on text representation methods for decision prediction based on graph structures, including causal graphs, element graphs, and knowledge graphs. Thirdly, this paper analyzes prediction methods based on graph deep learning, covering graph distillation operators, graph convolutional networks, gated graph neural networks, heterogeneous graph neural networks, graph attention networks, and graph reasoning networks. Fourthly, this paper introduces mainstream datasets, such as CAIL and ELAM, along with other datasets pertinent to judicial decision prediction, and arranges the evaluation metrics used in this domain. Finally, this paper envisions future research trends in legal judgment prediction, emphasizing the continuous innovation and application prospects of graph deep learning in this domain.

Key words: legal artificial intelligence, legal judgment prediction, graph deep learning

摘要： 司法判决预测旨在根据案件事实描述预测判决结果，如案件所涉及法条、罪名、刑期。近年来，图深度学习方法在法律人工智能领域取得了显著成效，能够有效地捕捉案件事实描述之间的复杂关系和法条之间的依赖关系。全面归纳并总结了图深度学习在司法判决预测任务上最新的研究成果。明确司法判决预测的任务定义，并回顾了早期的研究。重点关注基于图结构的司法判决预测的文本表示方法，具体包括因果图、要素图和知识图谱在本领域中的应用。分析图深度学习在司法判决预测中的相关方法，涵盖图蒸馏算子、图卷积网络、门控图神经网络、异构图神经网络、图注意力网络、图推理网络。介绍了CAIL、ELAM等主流数据集以及司法判决预测其他相关数据集，并整理了评估指标。展望司法判决预测的未来研究趋势，强调了图深度学习在此领域的持续创新和应用前景。

关键词: 法律人工智能, 司法判决预测, 图深度学习

ZHANG Yifei, LI Yanling, GE Fengpei. Review of Legal Judgment Prediction Based on Graph Deep Learning[J]. Journal of Frontiers of Computer Science and Technology, 2025, 19(8): 2024-2042.

张亦菲, 李艳玲, 葛凤培. 基于图深度学习的司法判决预测综述[J]. 计算机科学与探索, 2025, 19(8): 2024-2042.

References

[1] XIAO C J, ZHONG H X, GUO Z P, et al. CAIL2018: a large-scale legal dataset for judgment prediction[EB/OL]. [2024-07-03]. https://arxiv.org/abs/1807.02478.
[2] FENG Y, LI C Y, NG V. Legal judgment prediction: a survey of the state of the art[C]//Proceedings of the 31st International Joint Conference on Artificial Intelligence. Palo Alto: AAAI, 2022: 5461-5469.
[3] CUI J Y, SHEN X Y, WEN S C. A survey on legal judgment prediction: datasets, metrics, models and challenges[J]. IEEE Access, 2023, 11: 102050-102071.
[4] 彭亚男, 尹华, 贺敏伟. 中文法律条文推荐深度学习方法综述[J]. 软件导刊, 2023, 22(12): 245-252.
PENG Y N, YIN H, HE M W. Survey on Chinese law articles recommendation with deep learning[J]. Software Guide, 2023, 22(12): 245-252.
[5] NAGEL S. Weighting variables in judicial prediction[J]. Modern Uses of Logic in Law, 1960, 2(3): 93-97.
[6] LAWLOR R C. What computers can do: analysis and prediction of judicial decisions[J]. American Bar Association Journal, 1963, 49(4): 337-344.
[7] ULMER S S. Quantitative analysis of judicial processes: some practical and theoretical applications[J]. Law and Contemporary Problems, 1963, 28(1): 164.
[8] KEOWN R. Mathematical models for legal prediction[J]. Computer/Law Journal, 1980, 2(1): 829-830.
[9] LIU C L, HSIEH C D. Exploring phrase-based classification of judicial documents for criminal charges in Chinese[C]//Proceedings of the 16th International Symposium on Foundations of Intelligent Systems. Berlin, Heidelberg: Springer, 2006: 681-690.
[10] KATZ D M. Quantitative legal prediction-or-how I learned to stop worrying and start preparing for the data-driven future of the legal services industry[J]. Emory Law Journal, 2013, 62(4): 909-966.
[11] ALETRAS N, TSARAPATSANIS D, PREO?IUC-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: e93.
[12] ?ULEA O M, ZAMPIERI M,VELA M, et al. Predicting the law area and decisions of French supreme court cases[C]//Proceedings of the 2017 International Conference Recent Advances in Natural Language Processing, 2017: 716-722.
[13] SADIQ FAREED M M, RAZA A, ZHAO N, et al. Predicting divorce prospect using ensemble learning: support vector machine, linear model, and neural network[J]. Computational Intelligence and Neuroscience, 2022(1): 3687598.
[14] RAZA A, MUNIR K, ALMUTAIRI M, et al. Predicting employee attrition using machine learning approaches[J]. Applied Sciences, 2022, 12(13): 6424.
[15] LUO B F, FENG Y S, XU J B, et al. Learning to predict charges for criminal cases with legal basis[C]//Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing. Stroudsburg: ACL, 2017: 2727-2736.
[16] DEVLIN J, CHANG M W, LEE K, et al. BERT: pre-training of deep bidirectional transformers for language understanding[C]//Proceedings of the 2019 Conference of the North Amer-ican Chapter of the Association for Computational Linguistics. Stroudsburg: ACL, 2019: 4171-4186.
[17] BROWN T, MANN B, RYDER N, et al. Language models are few-shot learners[C]//Advances in Neural Information Processing Systems 33, 2020: 1877-1901.
[18] GE J D, HUANG Y Y, SHEN X Y, et al. Learning fine-grained fact-article correspondence in legal cases[J]. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2021, 29: 3694-3706.
[19] HUANG Y, SHEN X, LI C, et al. Dependency learning for legal judgment prediction with a unified text-to-text transformer[EB/OL]. [2024-07-03]. https://arxiv.org/abs/2112.06370.
[20] ZHONG H X, GUO Z P, TU C C, et al. Legal judgment prediction via topological learning[C]//Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing. Stroudsburg: ACL, 2018: 3540-3549.
[21] WEI D, LIN L. An external knowledge enhanced multi-label charge prediction approach with label number learning[EB/OL]. [2024-07-03]. https://arxiv.org/abs/1907.02205.
[22] YANG W M, JIA W J, ZHOU X J, et al. Legal judgment prediction via multi-perspective bi-feedback network[C]//Proceedings of the 28th International Joint Conference on Artificial Intelligence. Palo Alto: AAAI, 2019: 4085-4091.
[23] HARRIS Z S. Distributional structure[J]. WORD, 1954, 10(2/3): 146-162.
[24] BENGIO Y, DUCHARME R, VINCENT P, et al. A neural probabilistic language model[J]. Journal of Machine Learning Research, 2003, 3: 1137-1155.
[25] MARTINEAU J, FININ T. Delta TFIDF: an improved feature space for sentiment analysis[C]//Proceedings of the 2009 International AAAI Conference on Web and Social Media. Palo Alto: AAAI, 2009: 258-261.
[26] MIKOLOV T, CHEN K, CORRADO G, et al. Efficient estimation of word representations in vector space[EB/OL]. [2024-07-03]. https://arxiv.org/abs/1301.3781.
[27] PENNINGTON J, SOCHER R, MANNING C. GloVe: global vectors for word representation[C]//Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing. Stroudsburg: ACL, 2014: 1532-1543.
[28] SHAHBAZ M, SURESH L, REXFORD J, et al. Elmo: source routed multicast for public clouds[J]. IEEE/ACM Transactions on Networking, 2020, 28(6): 2587-2600.
[29] LECUN Y, BOSER B, DENKER J S, et al. Backpropagation applied to handwritten zip code recognition[J]. Neural Computation, 1989, 1(4): 541-551.
[30] ZAREMBA W, SUTSKEVER I, VINYALS O. Recurrent neural network regularization[EB/OL]. [2024-07-03]. https:// arxiv.org/abs/2310.11042.
[31] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Advances in Neural Information Processing Systems 30, 2017: 5998-6008.
[32] GEHRMANN S, DENG Y T, RUSH A. Bottom-up abstractive summarization[C]//Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing. Stroudsburg: ACL, 2018: 4098-4109.
[33] FILLMORE C J. The case for case[M]//Universals in linguistic theory. New York: Holt, Rinehart & Winston, 1968: 1-88.
[34] PEARL J. Causal diagrams for empirical research[J]. Biometrika, 1995, 82(4): 669-688.
[35] NIKOLIC M. Representation and reasoning: a causal model approach[D]. London: University College London, 2014.
[36] SO F. Modelling causality in law=Modélisation de la causalité en droit[EB/OL]. [2024-07-03]. https://hdl.handle.net/1866/25170.
[37] AL-FEDAGHI S. Diagrammatic modelling of causality and causal relations[EB/OL]. [2024-07-03]. https://arxiv.org/abs/ 2310.11042.
[38] TIKKA S, HELSKE J, KARVANEN J. Clustering and structural robustness in causal diagrams[EB/OL]. [2024-07-03]. https://arxiv.org/abs/2111.04513.
[39] 张虎, 张振, 范越, 等. 基于因果图分析的可解释司法判决预测方法研究[J]. 大数据, 2024, 10(2): 109-121.
ZHANG H, ZHANG Z, FAN Y, et al. Research on interpretable legal judgment prediction method based on causal graph analysis[J]. Big Data Research, 2024, 10(2): 109-121.
[40] CASTELLO J, REDMOND P, KUPER L. Inductive diagrams for causal reasoning[J]. Proceedings of the ACM on Programming Languages, 2024, 8: 529-554.
[41] ZHAO Q, GUO R D, FENG X W, et al. Research on a decision prediction method based on causal inference and a multi-expert FTOPJUDGE mechanism[J]. Mathematics, 2022, 10(13): 2281.
[42] IMBENS G W. Nonparametric estimation of average treatment effects under exogeneity: a review[J]. Review of Economics and Statistics, 2004, 86(1): 4-29.
[43] ANDRILLON A, PIRRACCHIO R, CHEVRET S. Performance of propensity score matching to estimate causal effects in small samples[J]. Statistical Methods in Medical Research, 2020, 29(3): 644-658.
[44] CAMPOS R, MANGARAVITE V, PASQUALI A, et al. YAKE! Keyword extraction from single documents using multiple local features[J]. Information Sciences, 2020, 509: 257-289.
[45] OGARRIO J M, SPIRTES P, RAMSEY J. A hybrid causal search algorithm for latent variable models[C]//Proceedings of the 8th International Conference on Probabilistic Graphical Models. Cambridge: MIT Press, 2016: 368-379.
[46] MADHUMITHA S, NADUVATH S. Graphs on groups in terms of the order of elements: a review[J]. Discrete Mathematics, Algorithms and Applications, 2024, 16(3): 2330003.
[47] 李瑾. 基于因果推理和法条信息融合的罪名预测方法研究[D]. 秦皇岛: 燕山大学, 2023.
LI J. Research on charge prediction method based on causal inference and law information fusion[D]. Qinhuangdao: Yanshan University, 2023.
[48] GUO Y, LI Y L, GE F P, et al. Legal judgment prediction via fine-grained element graphs and external knowledge[C]//Proceedings of the 2024 International Joint Conference on Neural Networks. Piscataway: IEEE, 2024: 1-8.
[49] FENSEL D, ?IM?EK U, ANGELE K, et al. Introduction: what is a knowledge graph[M]//Knowledge graphs: methodology, tools and selected use cases. Cham: Springer, 2020: 1-10.
[50] 黄治纲, 谢新强, 邢铁军, 等. 基于司法案例知识图谱的类案推荐[J]. 南京大学学报(自然科学), 2021, 57(6): 1053-1063.
HUANG Z G, XIE X Q, XING T J , et al. Case recommendation based on knowledge graph of judicial cases[J]. Journal of Nanjing University (Natural Science), 2021, 57(6): 1053-1063.
[51] 杜文源. 基于知识图谱的刑事案件判决预测算法研究[D]. 厦门: 厦门大学, 2020.
DU W Y. Research on judgment prediction algorithms of criminal cases based on domain knowledge graph[D]. Xiamen: Xiamen University, 2020.
[52] 王文同. 知识引导的司法判决预测研究[D]. 济南: 山东大学, 2022.
WANG W T. Knowledge-guided legal judgment prediction research[D]. Jinan: Shandong University, 2022.
[53] GAO S, SA R N, LI Y L, et al. How legal knowledge graph can help predict charges for legal text[C]//Proceedings of the 2023 International Conference on Neural Information Processing. Singapore: Springer, 2023: 408-420.
[54] SHI J M, GUO Q L, LIAO Y, et al. Legal-LM: knowledge graph enhanced large language models for law consulting[C]//Proceedings of the 20th International Conference on Advanced Intelligent Computing Technology and Applications. Singapore: Springer, 2024: 175-186.
[55] LI J, SUN A X, HAN J L, et al. A survey on deep learning for named entity recognition[J]. IEEE Transactions on Knowledge and Data Engineering, 2022, 34(1): 50-70.
[56] RABINER L R. A tutorial on hidden Markov models and selected applications in speech recognition[J]. Proceedings of the IEEE, 1989, 77(2): 257-286.
[57] LAFFERTY J, MCCALLUM A, PEREIRA F. Conditional random fields: probabilistic models for segmenting and labeling sequence data[C]//Proceedings of the 18th International Conference on Machine Learning, 2001: 282-289.
[58] HUANG Z H, XU W, YU K. Bidirectional LSTM-CRF models for sequence tagging[EB/OL]. [2024-07-06]. https://arxiv.org/abs/1508.01991.
[59] 刘博法. 以“四个坚持” 把握好“三个善于” 实现“高质效办好每一个案件”[J]. 中国检察官, 2024(9): 7-8.
LIU B F. Grasping “three masteries” with “four persistences” and realizing “running every case well with high quality and efficiency”[J]. The Chinese Procurators, 2024(9): 7-8.
[60] HINTON G. Distilling the knowledge in a neural network[EB/OL]. [2024-07-06]. https://arxiv.org/abs/1503.02531.
[61] YANG C, GUO Y X, XU Y, et al. Learning to distill graph neural networks[C]//Proceedings of the 16th ACM International Conference on Web Search and Data Mining. New York: ACM, 2023: 123-131.
[62] LIU Y, BO D, SHI C. Graph distillation with eigenbasis matching[C]//Proceedings of the 41st International Conference on Machine Learning, 2024: 30702-30717.
[63] XU N, WANG P H, 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. Stroudsburg: ACL, 2020: 3086-3095.
[64] TIAN Y J, PEI S C, ZHANG X L, et al. Knowledge distillation on graphs: a survey[J]. ACM Computing Surveys, 2025, 57(8): 1-16.
[65] KIPF T N, WELLING M. Semi-supervised classification with graph convolutional networks[EB/OL]. [2024-07-19]. https://arxiv.org/abs/1609.02907.
[66] 孙倩, 秦永彬, 黄瑞章, 等. 结合案件要素序列的罪名预测方法[J]. 大数据, 2021, 7(6): 30-40.
SUN Q, QIN Y B, HUANG R Z, et al. Charge prediction method combined with case elements sequence[J]. Big Data Research, 2021, 7(6): 30-40.
[67] GAN J Z, HU R Y, MO Y J, et al. Multigraph fusion for dynamic graph convolutional network[J]. IEEE Transactions on Neural Networks and Learning Systems, 2022, 35(1): 196-207.
[68] RUIZ L, GAMA F, RIBEIRO A. Gated graph recurrent neural networks[J]. IEEE Transactions on Signal Processing, 2020, 68: 6303-6318.
[69] ZHAO Q H, GAO T H, GUO N. LA-MGFM: a legal judgment prediction method via sememe-enhanced graph neural networks and multi-graph fusion mechanism[J]. Information Processing & Management, 2023, 60(5): 103455.
[70] WANG X, LIU N, HAN H, et al. Self-supervised heterogeneous graph neural network with co-contrastive learning[C]//Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. New York: ACM, 2021: 1726-1736.
[71] 栾厚蕴. 基于深度学习的法律判决预测系统的研究与实现[D]. 北京: 北京邮电大学, 2023.
LUAN H Y. Research and implementation of legal judgement prediction system based on deep learning[D]. Beijing: Beijing University of Posts and Telecommunications, 2023.
[72] BING R, YUAN G, ZHU M, et al. Heterogeneous graph neural networks analysis: a survey of techniques, evaluations and applications[J]. Artificial Intelligence Review, 2023, 56(8): 8003-8042.
[73] PAN W J, CHEN Y, LIU Z H, et al. Circumstance-aware graph neural network for legal judgment prediction[C]//Proceedings of the 2023 International Conference on Asian Language Processing. Piscataway: IEEE, 2023: 332-337.
[74] LIU Y F, WU Y Q, ZHANG Y T, et al. ML-LJP: multi-law aware legal judgment prediction[C]//Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2023: 1023-1034.
[75] TONG S X, YUAN J L, ZHANG P L, et al. Legal judgment prediction via graph boosting with constraints[J]. Information Processing & Management, 2024, 61(3): 103663.
[76] CHEN Y P, ROHRBACH M, YAN Z C, et al. Graph-based global reasoning networks[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2019: 433-442.
[77] LIU Y K, FENG S, WANG D L, et al. A graph reasoning network for multi-turn response selection via customized pre-training[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2021, 35(15): 13433-13442.
[78] JI Z, CHEN K X, HE Y Q, et al. Heterogeneous memory enhanced graph reasoning network for cross-modal retrieval[J]. Science China Information Sciences, 2022, 65(7): 172104.
[79] WANG J W, LE Y Q, CAO D, et al. Graph reasoning with supervised contrastive learning for legal judgment prediction[J]. IEEE Transactions on Neural Networks and Learning Systems, 2025, 36(2): 2801-2815.
[80] YU W J, SUN Z X, XU J, et al. Explainable legal case matching via inverse optimal transport-based rationale extraction[C]//Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2022: 657-668.
[81] FENG Y, LI C Y, NG V. Legal judgment prediction via event extraction with constraints[C]//Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics. Stroudsburg: ACL, 2022: 648-664.
[82] LI H T, SHAO Y Q, WU Y Y, et al. LeCaRDv2: a large-scale Chinese legal case retrieval dataset[C]//Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2024: 2251-2260.
[83] 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. Stroudsburg: ACL, 2018: 487-498.
[84] YANG Y M. An evaluation of statistical approaches to text categorization[J]. Information Retrieval, 1999, 1(1): 69-90.
[85] POWERS D M W. Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation[J]. Machine Learning Technologies, 2011, 2(1): 37-63.
[86] FISHER R A. The use of multiple measurements in taxonomic problems[J]. Annals of Eugenics, 1936, 7(2): 179-188.
[87] SONG H O, XIANG Y, JEGELKA S, et al. Deep metric learning via lifted structured feature embedding[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2016: 4004-4012.
[88] ROUGE L C Y. A package for automatic evaluation of summaries[C]//Proceedings of the 2004 Workshop on Text Summarization of ACL. Stroudsburg: ACL, 2004: 74-81.
[89] JACCARD P. The distribution of the flora in the alpine zone[J]. New Phytologist, 1912, 11(2): 37-50.
[90] 侯壹凡. 基于全局特征和局部特征融合的法条预测方法研究[D]. 湘潭: 湘潭大学, 2022.
HOU Y F. Research on the method of law prediction based on feature fusion of global features and local features[D]. Xiangtan: Xiangtan University, 2022.
[91] ZHANG Z P, LU S T, HUANG Z F, et al. ASGNN: graph neural networks with adaptive structure[EB/OL]. [2024-07-19]. https://arxiv.org/abs/2210.01002.
[92] RIBEIRO M T, SINGH S, GUESTRIN C. “Why should I trust you?”: explaining the predictions of any classifier[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2016: 1135-1144.
[93] WU Y Q, ZHOU S Y, LIU Y F, et al. Precedent-enhanced legal judgment prediction with LLM and domain-model collaboration[C]//Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing. Stroudsburg: ACL, 2023: 12060-12075.
[94] DENG C L, MAO K L, ZHANG Y Y, et al. Enabling discriminative reasoning in LLMs for legal judgment prediction[EB/OL]. [2024-11-03]. https://arxiv.org/abs/2407. 01964.