基于卷积神经网络的辅助分案方法研究

doi:10.11959/j.issn.2096-0271.2022020

大数据 ›› 2022, Vol. 8 ›› Issue (2): 145-157.doi: 10.11959/j.issn.2096-0271.2022020

基于卷积神经网络的辅助分案方法研究

敖绍林¹, 秦永彬¹^,², 黄瑞章¹^,², 陈艳平¹^,², 刘丽娟³, 郑庆华⁴, 陈昌恒⁵, 程少芬⁵

¹ 贵州大学计算机科学与技术学院，贵州贵阳 550025
² 公共大数据国家重点实验室，贵州贵阳 550025
³ 贵州师范学院，贵州贵阳 550025
⁴ 西安交通大学计算机科学与技术学院，陕西西安 710049
⁵ 贵州省高级人民法院，贵州贵阳 550081

出版日期:2022-03-15 发布日期:2022-03-01
作者简介:敖绍林（1995- ），男，贵州大学计算机科学与技术学院硕士生，主要研究方向为自然语言处理、文本分析
秦永彬（1980- ），男，博士，贵州大学计算机科学与技术学院教授，主要研究方向为大数据处理、云计算、文本挖掘
黄瑞章（1979- ），女，博士，贵州大学计算机科学与技术学院副教授，主要研究方向为信息检索、文本挖掘
陈艳平（1980- ），男，博士，贵州大学计算机科学与技术学院副教授，主要研究方向为人工智能、自然语言处理
刘丽娟（1980- ），女，贵州师范学院讲师，主要研究方向为法学与思想政治教育
郑庆华（1969- ），男，博士，西安交通大学计算机科学与技术学院教授，主要研究方向为多媒体远程教育、计算机网络安全
陈昌恒（1978- ），男，贵州省高级人民法院信息技术处处长、三级调研员，主要研究方向为司法审判应用
程少芬（1982- ），女，贵州省高级人民法院信息技术处应用推广科科长、四级调研员，主要研究方向为司法审判应用
基金资助:
国家自然科学基金资助项目(U1836205);国家自然科学基金资助项目(91746116);国家自然科学基金资助项目(62066007);国家自然科学基金资助项目(62066008);国家自然科学基金资助项目(62166007);贵州省科技重大专项计划([2017]3002);贵州省科学技术基金重点项目([2020]1Z055);贵州省研究生科研基金立项课题(YJSCXJH[2019]102)

Research on auxiliary division method based on convolutional neural network

Shaolin AO¹, Yongbin QIN¹^,², Ruizhang HUANG¹^,², Yanping CHEN¹^,², Lijuan LIU³, Qinghua ZHENG⁴, Changheng CHEN⁵, Shaofen CHENG⁵

¹ School of Computer Science and Technology, Guizhou University, Guiyang 550025, China
² State Key Laboratory of Public Big Data, Guiyang 550025, China
³ Guizhou Education University, Guiyang 550025, China
⁴ School of Computer Science and Technology, Xi’an Jiaotong University, Xi’an 710049, China
⁵ Guizhou Higher People’s Court, Guiyang 550081, China

Online:2022-03-15 Published:2022-03-01
Supported by:
The National Natural Science Foundation of China(U1836205);The National Natural Science Foundation of China(91746116);The National Natural Science Foundation of China(62066007);The National Natural Science Foundation of China(62066008);The National Natural Science Foundation of China(62166007);The Major Special Science and Technology Projects of Guizhou Province([2017]3002);The Key Projects of Science and Technology of Guizhou Province([2020]1Z055);Project of Guizhou Province Graduate Research Fund(YJSCXJH[2019]102)

摘要/Abstract

摘要：

法院系统中主要有人工指定分案和简单随机分案两种模式。这两种模式无法做到人案的自动匹配，存在金钱案、关系案等弊端。目前分案方法的相关研究主要存在法官表示和案件匹配两个难点。结合法官历史审判数据，在法官表示中融合法官擅长的审判领域，提出一种融合审判质量的法官表示方法。然后，通过卷积神经网络学习案件表示和法官表示中不同粒度的抽象语义特征向量，计算案件和多个法官的特征向量间的余弦相似度，用向量相似度表示案件与法官的匹配度，输出匹配值较高的前N个法官作为案件的推荐法官。在贵州省某法院真实数据下进行实验，结果表明该方法推荐法官的正确率比传统方法高80%。

关键词: 文本表示, 卷积神经网络, 智能分案, 智慧法院

Abstract:

The court system mainly has two modes: manual designated division and simple random division.The above method cannot achieve automatic matching of persons and cases, and there are drawbacks such as money cases and relationship cases.At present, the research on division method mainly has two difficulties: judge’s representation and case matching.Combining the judge’s historical trial data, the judge’s expertise in the judge’s representation was integrated, and a judge representation method that integrates the quality of the trial was proposed.Then, the abstract semantic feature vectors of different granularities in the case representation and the judge representation were learned through the convolutional neural network, the cosine similarity between the case and the feature vectors of multiple judges was calculated, and vector similarity was used to indicate the matching degree between the case and the judge, the top N judges with high matching value were output as recommended judges for the case.Experiments with real data from a court in Guizhou Province, and the results show that the accuracy of the method for recommending judges is 80% higher than the traditional method.

Key words: text representation, convolutional neural network, smart division, smart court

中图分类号:

TP183

敖绍林, 秦永彬, 黄瑞章, 陈艳平, 刘丽娟, 郑庆华, 陈昌恒, 程少芬. 基于卷积神经网络的辅助分案方法研究[J]. 大数据, 2022, 8(2): 145-157.

Shaolin AO, Yongbin QIN, Ruizhang HUANG, Yanping CHEN, Lijuan LIU, Qinghua ZHENG, Changheng CHEN, Shaofen CHENG. Research on auxiliary division method based on convolutional neural network[J]. Big Data Research, 2022, 8(2): 145-157.

图/表 8

图1

图2

表1

表2

表3

表4

表5

表6

参考文献 24

[1]	秦永彬, 冯丽, 陈艳平 ,等. “智慧法院”数据融合分析与集成应用[J]. 大数据, 2019,5(3): 35-46.
	QIN Y B , FENG L , CHEN Y P ,et al. “Intelligent Court” data fusion analysis and integrated application[J]. Big Data Research, 2019,5(3): 35-46.
[2]	SALTON G , BUCKLEY C . Termweighting approaches in automatic text retrieval[J]. Information Processing ＆Management, 1988,24(5): 513-523.
[3]	DEERWESTER S , DUMAIS S T , FURNAS G W ,et al. Indexing by latent semantic analysis[J]. Journal of the American Society for Information Science, 1990,41(6): 391-407.
[4]	BLEI D M , NG A Y , JORDAN M I . Latent Dirichlet allocation[J]. Journal of Machine Learning Research, 2001,3: 601-608.
[5]	BENGIO Y , DUCHARME R , VINCENT P ,et al. A neural probabilistic language model[J]. Journal of Machine Learning Research, 2003,3: 1137-1155.
[6]	MIKOLOV T , CHEN K , CORRADO G ,et al. Efficient estimation of word representations in vector space[J]. arXiv preprint,2013,arXiv:1301.3781.
[7]	MIKOLOV T , SUTSKEVER I , CHEN K ,et al. Distributed representations of words and phrases and their compositionality[C]// Proceedings of the 26th International Conference in Neural Information Processing Systems.Red Hook:Curran Associates Inc. , 2013: 3111-3119.
[8]	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:Association for Computational Linguistics, 2014: 1532-1543.
[9]	BOJANOWSKI P , GRAVE E , JOULIN A ,et al. Enriching word vectors with subword information[J]. Transactions of the Association for Computational Linguistics, 2017,5: 135-146.
[10]	PETERS M E , NEUMANN M , IYYER M ,et al. Deep contextualized word representations[J]. arXiv preprint,2018,arXiv:1802.05365.
[11]	DEVLIN J , CHANG M W , LEE K ,et al. BERT:pre-training of deep bidirectional transformers for language understanding[J]. arXiv preprint,2018,arXiv:1810.04805.
[12]	KIM Y . Convolutional neural networks for sentence classification[J]. arXiv preprint,2014,arXiv:1408.5882.
[13]	冯兴杰, 张乐, 曾云泽 . 基于多注意力CNN的问题相似度计算模型[J]. 计算机工程, 2019,45(9): 284-290.
	FENG X J , ZHANG L , ZENG Y Z . Question similarity calculation model based on multiattention CNN[J]. Computer Engineering, 2019,45(9): 284-290.
[14]	CHIU J P C , NICHOLS E . Named entity recognition with bidirectional LSTMCNNs[J]. Transactions of the Association for Computational Linguistics, 2016,4: 357-370.
[15]	孙国荣 . 智能分案:兼顾公平与效率[J]. 人民法治, 2018(2): 49-51.
	SUN G R . Smart division:taking into account fairness and efficiency[J]. People·Rule of Law, 2018(2): 49-51.
[16]	王崇璋 . 法院案件分配标准体系实证研究:以HB市中基层法院为例[D]. 淮北:淮北师范大学, 2018.
	WANG C Z . An empirical study on the standard system of court case allocation-take the example of the HB municipal intermediate and basic courts[D]. Huaibei:Huaibei Normal University, 2018.
[17]	王建华, 张宁 . 美国联邦法院民事案件的立案规则[N]. 人民法院报, 2015-10-23(8).
	WANG J H , ZHANG N . Rules for filing civil cases in the U.S.Federal Court[N]. People’s Court Daily, 2015-10-23(8).
[18]	兰世民, 兰馨, 缪新森 . 法院分案若干问题研究[J]. 法律适用, 2012(6): 95-100.
	LAN S M , LAN X , MIAO X S . Research on several Issues of court division[J]. Journal of Law Application, 2012(6): 95-100.
[19]	贾炜, 常鹏, 石鹏 . 一种实现法院分案的方法及装置:CN107665212A[P]. 2018-02-06.
	JIA W , CHANG P , SHI P . Method and device for case allocation of court:CN107665212A[P]. 2018-02-06.
[20]	顾钊铨, 方滨兴, 韩伟红 ,等. 一种法院案件智能化分案辅助方法及系统:CN109872052A[P]. 2019-06-11.
	GU Z Q , FANG B X , HAN W H ,et al. Court case intelligent case division auxiliary method and system:CN109872052A[P]. 2019-06-11.
[21]	陈芳序 . 法官工作量均等视角下法院随机分案系统的检视与重构:以A法院刑事分案为切入点[J]. 海峡法学, 2019,21(2): 87-95.
	CHEN F X . Inspection and reconstruction of the court random division system from perspective of equal workload for judges-based on criminal division of court A[J]. Cross Strait Legal Science, 2019,21(2): 87-95.
[22]	王小新 . 法院分案系统的检视与重构:以X法院刑事案件分配为例[J]. 法律适用, 2016(4): 112-116.
	WANG X X . Review and reconstruction of the court division system:taking X court criminal case distribution as an example[J]. Journal of Law Application, 2016(4): 112-116.
[23]	孙晓东 . 中国司法评估制度完善研究[J]. 广东社会科学, 2018(6): 231-242.
	SUN X D . An analysis on improvement of judicial performance evaluation system of China[J]. Social Sciences in Guangdong, 2018(6): 231-242.
[24]	LAN Z Z , CHEN M D , GOODMAN S ,et al. ALBERT:a lite BERT for selfsupervised learning of language representations[J]. arXiv preprint,2019,arXiv:1909.11942.

数据集类别	案件数/个	三元组数/个
训练集	1 238	21 046
验证集	152	2 584
测试集	156	2 652

N	正确率
1	86.54%
3	96.15%
5	98.07%
7	98.72%

分案方法	正确率
分案方法	Top1	Top3	Top5	Top7
均衡分案	5.85%	18.72%	30.77%	43.07%
摇号分案	5.40%	18.02%	27.27%	35.85%
本文方法	86.54%	96.15%	98.07%	98.72%

分案方法	正确率
分案方法	Top1	Top3	Top5	Top7
TF-IDF+cosine	1.28%	24.36%	57.05%	66.03%
BM25+cosine	16.67%	43.59%	66.67%	90.38%
本文方法	86.54%	96.15%	98.07%	98.72%

分案方法	正确率
分案方法	Top1	Top3	Top5	Top7
Tri-BiLSTM	49.36%	75.64%	88.46%	96.15%
Tri-BERT	78.21%	91.67%	96.15%	97.44%
Tri-ALBERT	76.92%	94.23%	97.44%	98.72%
本文方法	86.54%	96.15%	98.07%	98.72%

基于卷积神经网络的辅助分案方法研究

Research on auxiliary division method based on convolutional neural network

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 24

相关文章 4

Metrics

推荐阅读 0

参数ε	正确率
参数ε	Top1	Top3	Top5	Top7
0.1	79.49%	94.23%	97.44%	98.08%
0.5	82.69%	92.95%	97.44%	99.36%
1.0	86.54%	96.15%	98.07%	98.72%

[1]	孙倩, 秦永彬, 黄瑞章, 刘丽娟, 陈艳平. 结合案件要素序列的罪名预测方法[J]. 大数据, 2021, 7(6): 30-40.
[2]	王丽会, 秦永彬. 深度学习在医学影像中的研究进展及发展趋势[J]. 大数据, 2020, 6(6): 83-104.
[3]	宋婷, 陈战伟, 杨海峰. 基于分层注意力网络的方面情感分析[J]. 大数据, 2020, 6(5): 82-91.
[4]	秦永彬, 冯丽, 陈艳平, 黄瑞章, 刘于雷, 丁红发. “智慧法院”数据融合分析与集成应用[J]. 大数据, 2019, 5(3): 35-46.