摘要
当前法院案件数量持续增加,而法官员额固定不变。为提高法官的办案效率,利用计算机相关技术帮助或辅助法官对案件进行推理分析,是解决案件繁重的有效途径。为解决上述问题,在详细分析现有神经网络模型及相关算法的基础上,使用Word2Vec对案件样本进行向量化处理,利用长短时记忆算法创建了案件智能推理辅助分析模型。首先介绍系统的各个组成部分及其实现逻辑和算法处理过程,而后对案件样本进行训练生成智能推理辅助分析模型,最后参考案件笔录对案件进行辅助分析得出推荐结论。系统结合房屋租赁合同纠纷案例对案件推理分析结论被采纳的情况进行了评估,通过实际使用和测试,案件推理分析结论被法官采纳的有效率达到80%左右,有效地提高了法官的办案效率。
With the continuous increase in the number of court cases and the fixed number of judge positions,in order to reduce the workload of judges and improve their efficiency in handling cases,the only effective way to solve the heavy workload of cases is the use of computer related technology to assist judges in thinking.To solve the above problems,based on the detailed analysis of existing neural network models and related algorithms,we use Word2Vec to vectorization the case samples,and use long short-term memory(LSTM)algorithm to create a case intelligent reasoning auxiliary analysis model.Firstly,the various components of the system,their implementation logic,and algorithm processing are introduced.Then,the case samples are trained to generate an intelligent reason-assisting analysis model,and combined with the case records,the case is analyzed to draw recommended conclusions.The system evaluated the adoption of case reasoning and analysis conclusions in combination with housing lease contract disputes.Through the practical use and testing,the effectiveness rate of the case reasoning and the analysis conclusions being adopted by judges achieve about 80%,effectively improving the efficiency of judges in handling cases and other issues.
作者
赵永翼
魏晓东
ZHAO Yongyi;WEI Xiaodong(Software College,Shenyang Normal University,Shenyang 110034,China;Law School,Liaoning University,Shenyang 110136,China)
出处
《沈阳师范大学学报（自然科学版）》
CAS
2023年第4期310-315,共6页
Journal of Shenyang Normal University:Natural Science Edition
基金
中国法学会法学研究项目(CLS(2020)ZX043)。
关键词
长短时记忆算法
案件智能推理
神经网络
数据挖掘
long short-term memory algorithm
case intelligent reasoning
neural network
data mining
