大数据证据相关性的四维审查与规则建构  

Four Dimensional Interpretation and Rule Construction of Relevance Review of Big Data Evidence

作　　者：陈在上 任逸飞 Chen Zaishang;Ren Yifei(School of Criminal Justice,Henan University of Economics and Law,Zhengzhou,China;School of law,Henan University of Economics and Law,Zhengzhou,China)

机构地区：[1]河南财经政法大学刑事司法学院,河南郑州450046 [2]河南财经政法大学法学院,河南郑州450046

出　　处：《社会科学论坛》2025年第2期112-123,共12页Tribune of Social Sciences

基　　金：河南省哲学社会科学规划办年度项目《审判中心主义视阈下律师辩护权实质化研究》阶段性成果,项目编号:2021BFX004。

摘　　要：大数据证据作为数字时代的新兴产物,频现于当前司法实践中,但囿于海量数据与复杂技术,致使传统证据相关性审查规则失灵。鉴于此,需从载体、数据、算法及结论四个维度建构大数据证据相关性审查规则。大数据证据的载体需拆解为身份、行为与介质,从形式上逐一审查相关性。基础数据的选取需同时满足数量、质量与力量要求,避免不相关数据的干扰。算法模型的构建与选取需评估合理性、稳健性与透明性,确保分析过程的科学适当。结论的审查需以印证方法为基础,引入概率方法与故事方法,从不同角度实现分析结果的全面审查。As an emerging product of the digital age,big data evidence frequently appears in current judicial practice.However,due to the limitations of massive data and complex technology,traditional rules for reviewing the relevance of evidence have become ineffective.In view of this,it is necessary to construct rules for reviewing the relevance of big data evidence from four dimensions:carrier,data,algorithm,and conclusion.The carrier of big data evidence needs to be broken down into identity,behavior,and medium,and the correlation should be examined one by one from a formal perspective.The selection of basic data should meet the requirements of quantity,quality,and strength simultaneously,avoiding interference from irrelevant data.The construction and selection of algorithm models need to evaluate their rationality,robustness,and transparency to ensure the scientific appropriateness of the analysis process.The review of conclusions should be based on verification methods,introducing probability methods and story methods to achieve a comprehensive review of analysis results from different perspectives.

关 键 词：大数据证据 证据审查 相关性 整体主义 印证方法 

分 类 号：DF73[政治法律—诉讼法学]