证言可信性评价的贝叶斯网络模型  

Bayesian Network Model for Credibility Evaluation of Testimony

作　　者：杜文静 DU Wenjing(Wenbo College,East China University of Political Science and Law,Shanghai 201620,China)

机构地区：[1]华东政法大学文伯书院,上海201620

出　　处：《山西大学学报（哲学社会科学版）》2024年第6期40-49,共10页Journal of Shanxi University(Philosophy and Social Science Edition)

基　　金：国家社科基金一般项目“司法证明的逻辑研究”(23BZX127)。

摘　　要：证言可信性评价包含两个研究维度:一是可信性的结构;二是可信性的程度。通过辨析证言报告、证言来源及证言内容等概念,结合对可靠性、似真性和可信性等要素的理解,有助于澄清可信性的逻辑结构。证言来源的可靠性和证言内容的似真性是影响证言可信性的直接因素,其中可靠性又可细化为诚实性、叙述能力、感知能力以及记忆能力四要素。基于概率理论和贝叶斯网络,可以构建出三个评价证言可信性的模型,它们按照细粒度和准确性从小到大依次递增。这些模型不仅清晰地展现了证言可信性评价的结构,还提供了一个解决方案,用以量化表达可靠性和似真性的实质审查结果,进而形成证言的综合可信度。The evaluation of the credibility of testimony encompasses two research dimensions:the structure of credibility and the degree of credibility.By discerning concepts such as testimony reports,sources of testimony,and content of testimony,and combining an understanding of elements such as reliability,plausibility and credibility,we can clarify the logical structure of credibility.The reliability of the source of testimony and the plausibility of the content of testimony are direct factors affecting the credibility of testimony,in which reliability can be further refined into four elements:honesty,narrative ability,perception ability and memory ability.Based on probability theory and Bayesian networks,we can construct three models for evaluating the credibility of testimony,which increase sequentially in terms of granularity and accuracy.These models not only clearly demonstrate the structure of the credibility of testimony,but also provide a solution for the quantitative expression of the essential examination results of reliability and plausibility,thereby forming a comprehensive degree of credibility of testimony.

关 键 词：证言 可信性 概率 贝叶斯网络 

分 类 号：D90-051[政治法律—法学理论]