Please wait a minute...
浙江大学学报(工学版)  2024, Vol. 58 Issue (7): 1357-1365    DOI: 10.3785/j.issn.1008-973X.2024.07.005
计算机与控制工程     
基于法条知识的事理型类案检索方法
李林睿1(),王东升2,*(),范红杰2
1. 浙江大学 光华法学院,浙江 杭州 310008
2. 中国政法大学 法治信息管理学院,北京 102249
Fact-based similar case retrieval methods based on statutory knowledge
Linrui LI1(),Dongsheng WANG2,*(),Hongjie FAN2
1. Guanghua Law School, Zhejiang University, Hangzhou 310008, China
2. School of Information Management for Law, China University of Political Science and Law, Beijing 102249, China
 全文: PDF(814 KB)   HTML
摘要:

现有类案检索研究忽略了模型应当蕴含的法律逻辑,无法适应实际应用中案件相似标准的要求;类案检索任务的中文数据集较少,难以满足研究需求现状. 为此提出基于法律逻辑、有较强可解释性的类案检索模型,构建以谓语动词为基础的案件事理图谱. 将各类罪名对应的法条知识融入所提模型,将提取的不同要素输入以神经网络为基础的评分器以实现准确、高效的类案检索. 构建针对类案检索任务、以易混淆罪名组为主要检索案由的Confusing-LeCaRD数据集,所提模型在LeCaRD数据集和Confusing-LeCaRD数据集上的归一化折损累计增益分别为90.95%和94.64%,在各项指标上均优于TF-IDF、BM25和BERT-PLI模型.

关键词: 类案检索法条知识法律逻辑事理图谱深度学习    
Abstract:

Existing research on the retrieval task of similar cases ignores the legal logic that the model should imply, and cannot adapt to the requirements of case similarity criteria in practical applications. Few datasets in Chinese for case retrieval tasks are difficult to meet the research needs. A similar case retrieval model was proposed based on legal logic and strong interpretability, and a case event logic graph was constructed based on predicate verbs. The statutory knowledge corresponding to various crimes was integrated into the proposed model, and the extracted elements were input to a neural network-based scorer to realize the task of case retrieval accurately and efficiently. A Confusing-LeCaRD dataset was built for the case retrieval task with a confusing group of charges as the main retrieval causes. Experiments show that the normalized discounted cumulative gain of the proposed model on the LeCaRD dataset and Confusing-LeCaRD dataset was 90.95% and 94.64%, and the model was superior to TF-IDF, BM25 and BERT-PLI in all indicators.

Key words: similar case retrieval    statutory knowledge    legal logic    event logic graph    deep learning
收稿日期: 2023-06-21 出版日期: 2024-07-01
CLC:  TP 391:TP 181  
基金资助: 长沙市科技重大专项项目(kh2202006);中国政法大学科研创新项目(24KYGH013);中央高校基本科研业务费专项资金.
通讯作者: 王东升     E-mail: lilinrui1412@163.com;wangdsh@cupl.edu.cn
作者简介: 李林睿(2001—),女,硕士生,从事数字法治研究. orcid.org/0009-0003-4988-7618. E-mail:lilinrui1412@163.com
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
作者相关文章  
李林睿
王东升
范红杰

引用本文:

李林睿,王东升,范红杰. 基于法条知识的事理型类案检索方法[J]. 浙江大学学报(工学版), 2024, 58(7): 1357-1365.

Linrui LI,Dongsheng WANG,Hongjie FAN. Fact-based similar case retrieval methods based on statutory knowledge. Journal of ZheJiang University (Engineering Science), 2024, 58(7): 1357-1365.

链接本文:

https://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2024.07.005        https://www.zjujournals.com/eng/CN/Y2024/V58/I7/1357

图 1  类案检索模型的整体架构
罪名要件式定义
虐待部属滥用职权,虐待部署,致人重伤,造成其他严重后果
妨碍安全驾驶对行驶中的交通工具的驾驶人员使用暴力,抢控驾驶操纵装置,干扰公共交通工具正常行驶,危及公共安全
重大责任事故违反有关安全管理的规定,发生重大伤亡事故,造成其他严重后果
表 1  法条知识库的部分条文
图 2  案件事理图谱的构建过程图
图 3  向量化处理和评分器
要件事实案情事实相似度
相似相似3
相似不相似2
不相似相似1
不相似不相似0
表 2  LeCaRD数据集的标注规则
罪名要件
危险驾驶在道路上驾驶机动车追逐竞驶,醉酒驾驶机动车,从事校车业务或者旅客运输严重超过定额乘员载客,从事校车业务或者旅客运输严重超过规定时速行驶,违反化学品安全管理规定运输危险化学品
交通肇事违反交通运输管理法规,致人重伤,致人死亡,造成重大公私财产损失
表 3  危险驾驶罪和交通肇事罪的要件
编号罪名案情
1诈骗2009年3月20日,被告甲为实施诈骗活动,通过中介注册成立了一家公司,并通过网络招聘了五名员工······
9抢劫2021年5月2日,被告甲见乙作为老年人独自一人行走在路上,便冲过去夺走乙手中的布包······
表 4  Confusing-LeCaRD数据集检索案件的数据结构
编号罪名案情相似度等级
1307抢劫2018年8月27日,甲骑着电动车在路上行驶,
被告乙驾驶摩托车从旁快速经过,乙车后座
上的被告丙夺走甲的手提包······
3
2167盗窃2019年9月25日,被告甲见乙房屋门未关,便
偷偷潜入乙家中窃取一部手机和两百元现金······
0
表 5  Confusing-LeCaRD数据集候选案件的数据结构
参数数值参数数值
Learning_rate2×10?4Max_len192
Batch_size1Hidden_size128
Weight_decay0.005Key_fact_threshold0.15
表 6  模型训练参数
模型NDCG@5NDCG@10NDCG@20NDCG@30
TF-IDF67.2373.4678.3683.40
BM2572.0873.8481.9787.41
BERT-PLI83.1985.6691.0191.17
本研究92.0494.6492.6091.51
表 7  不同模型基于Confusing-LeCaRD数据集的评测结果
消除的
模块
NDCG@5NDCG@10NDCG@20NDCG@30
法条知识88.00
(↓3.64)
90.34
(↓4.30)
88.22
(↓4.38)
89.82
(↓1.69)
事理图谱88.53
(↓3.51)
89.08
(↓5.56)
87.97
(↓4.63)
88.26
(↓3.25)
表 8  基于Confusing-LeCaRD数据集的消融实验结果
案件编号罪名要件事实基本事实
+17011交通肇事{死亡,造成肺挫伤,受伤,相撞,休克}驾驶→倒车→相撞→受伤→抢救无效→死亡→驾驶→未注意瞭望→
造成受伤→导致休克→经抢救无效→死亡→达成赔偿
+17189交通肇事{损伤,受伤,死亡,肇事,有死亡,损伤,
安全法违反}
驾驶→行驶→适逢→驾驶→行驶→相撞→受伤→拨打→送往救治→
肇事→治疗→死亡→损伤→死亡→达成赔偿→接受赔偿
+717盗窃{财物盗窃,占有,盗得,实施盗窃,入户}乘坐→驾驶→到达→盗窃→进入→盗得→进入→盗得→占有→入户
盗窃→追缴违法所得
表 9  检索案例和候选案例的特征信息
1 BHATTACHARYA P, GHOSH K, PAL A, et al. Methods for computing legal document similarity: a comparative study [EB/OL]. (2020-04-26)[2023-08-10]. https://arxiv.org/pdf/2004.12307.
2 WAGH R S, ANAND D Legal document similarity: a multi-criteria decision-making perspective[J]. PeerJ Computer Science, 2020, 6: e262
doi: 10.7717/peerj-cs.262
3 TRAN V, NGUYEN M L, SATOH K. Building legal case retrieval systems with lexical matching and summarization using a pre-trained phrase scoring model [C]// Proceedings of the Seventeenth International Conference on Artificial Intelligence and Law . [S.l.]: Association for Computing Machinery, 2019: 275–282.
4 JIANG J Y, ZHANG M, LI C, et al. Semantic text matching for long-form documents [C]// The World Wide Web Conference . [S.l.]: Association for Computing Machinery, 2019: 795–806.
5 SHAO Y, MAO J, LIU Y, et al. BERT-PLI: modeling paragraph-level interactions for legal case retrieval [C]// International Joint Conference on Artificial Intelligence . Yokohama: [s.n.], 2020: 3501–3507.
6 ALI B, MORE R, PAWAR S, et al. Prior case retrieval using evidence extraction from court judgements [C]// The Fifth Workshop on Automated Semantic Analysis of Information in Legal Text . São Paulo: [s.n.], 2021: 1–11.
7 MA Y, SHAO Y, WU Y, et al. LeCaRD: a legal case retrieval dataset for Chinese law system [C]// Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval . [S.l.]: Association for Computing Machinery, 2021: 2342–2348.
8 YOSHIOKA M, KANO Y, KIYOTA N, et al. Overview of japanese statute law retrieval and entailment task at COLIEE-2018[C]// Proceedings of the Twelfth International Workshop on Juris-Informatics . Yokohama: [s.n.], 2018: 117–128
9 赵京胜, 宋梦雪, 高祥, 等 自然语言处理中的文本表示研究[J]. 软件学报, 2022, 33 (1): 102- 128
ZHAO Jingsheng, SONG Mengxue, GAO Xiang, et al Research on text representation in natural language processing[J]. Journal of Software, 2022, 33 (1): 102- 128
10 WEI L, ZHOU C, SU R, et al PEPred-Suite: improved and robust prediction of therapeutic peptides using adaptive feature representation learning[J]. Bioinformatics, 2019, 35 (21): 4272- 4280
doi: 10.1093/bioinformatics/btz246
11 LEE H Y, HUANG J B, SINGH M, et al. Unsupervised representation learning by sorting sequences [C]// Proceedings of the IEEE International Conference on Computer Vision . Venice: IEEE, 2017: 667–676.
12 SUK H I, SHEN D. Deep learning-based feature representation for AD/MCI classification [C]// International Conference on Medical Image Computing and Computer Assisted Intervention . [S.l.]: Springer. 2013: 583–590.
13 李松, 舒世泰, 郝晓红, 等 融合文本描述和层次类型的知识表示学习方法[J]. 浙江大学学报: 工学版, 2023, 57 (5): 911- 920
LI Song, SHU Shitai, HAO Xiaohong, et al Knowledge representation learning method integrating textual description and hierarchical type[J]. Journal of Zhejiang University: Engineering Science, 2023, 57 (5): 911- 920
14 ZHONG H, ZHOU J, QU W, et al. An element-aware multi-representation model for law article prediction [C]// Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing . [S.l.]: Association for Computational Linguistics, 2020: 6663–6668.
15 LI L, SHI X, DING Y, et al. Event logic graph construction for event mining [C]// Journal of Physics: Conference Series . Beijing: IOP, 2021, 2037: 012135.
16 DING X, LI Z, LIU T, et al. ELG: an event logic graph [EB/OL]. (2019-08-07)[2023-08-10]. https://arxiv.org/pdf/1907.08015.
17 DU L, DING X, XIONG K, et al. ExCAR: event graph knowledge enhanced explainable causal reasoning [C]// Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing . [S.l.]: Association for Computational Linguistics, 2021: 2354–2363.
18 WANG Z. Legal element-oriented modeling with multi-view contrastive learning for legal case retrieval [C]// International Joint Conference on Neural Networks . Padua: IEEE, 2022: 1–10.
19 SHAO Y, WU Y, LIU Y, et al Understanding relevance judgments in legal case retrieval[J]. ACM Transactions on Information Systems, 2023, 41 (3): 1- 32
20 HU W, ZHAO S, ZHAO Q, et al BERT_LF: a similar case retrieval method based on legal facts[J]. Wireless Communications and Mobile Computing, 2022, 2022: 1- 9
21 NIGAM S K, GOEL N, BHATTACHARYA A. Nigam@COLIEE-22: legal case retrieval and entailment using cascading of lexical and semantic-based models [C]// New Frontiers in Artificial Intelligence . [S.l.]: Springer, 2023: 96–108.
22 TAN M, JIANG J, DAI B T A BERT-based two-stage model for Chinese chengyu recommendation[J]. ACM Transactions on Asian and Low-Resource Language Information Processing, 2021, 20 (6): 1- 18
[1] 吴书晗,王丹,陈远方,贾子钰,张越棋,许萌. 融合注意力的滤波器组双视图图卷积运动想象脑电分类[J]. 浙江大学学报(工学版), 2024, 58(7): 1326-1335.
[2] 宋娟,贺龙喜,龙会平. 基于深度学习的隧道衬砌多病害检测算法[J]. 浙江大学学报(工学版), 2024, 58(6): 1161-1173.
[3] 钟博,王鹏飞,王乙乔,王晓玲. 基于深度学习的EEG数据分析技术综述[J]. 浙江大学学报(工学版), 2024, 58(5): 879-890.
[4] 魏翠婷,赵唯坚,孙博超,刘芸怡. 基于改进Mask R-CNN与双目视觉的智能配筋检测[J]. 浙江大学学报(工学版), 2024, 58(5): 1009-1019.
[5] 宦海,盛宇,顾晨曦. 基于遥感图像道路提取的全局指导多特征融合网络[J]. 浙江大学学报(工学版), 2024, 58(4): 696-707.
[6] 罗向龙,王亚飞,王彦博,王立新. 基于双向门控式宽度学习系统的监测数据结构变形预测[J]. 浙江大学学报(工学版), 2024, 58(4): 729-736.
[7] 宋明俊,严文,邓益昭,张俊然,涂海燕. 轻量化机器人抓取位姿实时检测算法[J]. 浙江大学学报(工学版), 2024, 58(3): 599-610.
[8] 钱庆杰,余军合,战洪飞,王瑞,胡健. 基于DL-BiGRU多特征融合的注塑件尺寸预测方法[J]. 浙江大学学报(工学版), 2024, 58(3): 646-654.
[9] 姚鑫骅,于涛,封森文,马梓健,栾丛丛,沈洪垚. 基于图神经网络的零件机加工特征识别方法[J]. 浙江大学学报(工学版), 2024, 58(2): 349-359.
[10] 孙雪菲,张瑞峰,关欣,李锵. 强化先验骨架结构的轻量型高效人体姿态估计[J]. 浙江大学学报(工学版), 2024, 58(1): 50-60.
[11] 郑超昊,尹志伟,曾钢锋,许月萍,周鹏,刘莉. 基于时空深度学习模型的数值降水预报后处理[J]. 浙江大学学报(工学版), 2023, 57(9): 1756-1765.
[12] 杨哲,葛洪伟,李婷. 特征融合与分发的多专家并行推荐算法框架[J]. 浙江大学学报(工学版), 2023, 57(7): 1317-1325.
[13] 李云红,段姣姣,苏雪平,张蕾涛,于惠康,刘杏瑞. 基于改进生成对抗网络的书法字生成算法[J]. 浙江大学学报(工学版), 2023, 57(7): 1326-1334.
[14] 权巍,蔡永青,王超,宋佳,孙鸿凯,李林轩. 基于3D-ResNet双流网络的VR病评估模型[J]. 浙江大学学报(工学版), 2023, 57(7): 1345-1353.
[15] 赵嘉墀,王天琪,曾丽芳,邵雪明. 基于GRU的扑翼非定常气动特性快速预测[J]. 浙江大学学报(工学版), 2023, 57(6): 1251-1256.