基于图谱融合的人工智能司法数据库构建研究

Research on Artificial Intelligence Judicial Database Construction Based on Graph Fusion

传播影响力
本库下载频次:
本库浏览频次:
CNKI下载频次:0

归属学者:

朱福勇

作者:

朱福勇 ;刘雅迪 ;高帆 ;王凯

摘要:

司法与人工智能的融合是助力法院审判体系和审判能力智能化的基础。人工智能技术的飞速发展为构建司法知识库提供了必要的支持与帮助。当前的诉讼知识库是以树状层次化结构组织实体和节点关系,知识库中的实体关系主要表现为概念之间简单的"类属关系""同义词关系"。由于司法领域诸如证据链条、事理关系、法律判定规则等知识结构的复杂性,无法简单地组织、存储和应用数据结构和存储技术。是以,需要采用知识图谱和事理图谱相融合的技术,将司法领域中的专业术语和法律关系进行程序化表达,把标准化术语存储到证据要素知识模型中,以证据要素模板表达证据要素知识库,结合证据链条模型和证据规则模型的推理算法,从证据有效性、诉讼时效性、当事人法律行为规范性、诉讼请求合理性等方面对诉讼风险的提醒,以实现持久化的存储、事实推理和风险预测。

语种:

中文

出版日期:

2019-11-29

学科:

情报学; 民商法学

提交日期

2020-01-09

引用参考

朱福勇;刘雅迪;高帆;王凯. 基于图谱融合的人工智能司法数据库构建研究[J]. 扬州大学学报(人文社会科学版),2019(06):89-96.

  • dc.title
  • 基于图谱融合的人工智能司法数据库构建研究
  • dc.contributor.author
  • 朱福勇;刘雅迪;高帆;王凯
  • dc.contributor.author
  • ZHU Fuyong;LIU Yadi;GAO Fan;WANG Kai;Southwest University of Political Science and Law;Zhongjing Baicheng Technology Co, Ltd
  • dc.contributor.affiliation
  • 西南政法大学人工智能法学院;中经柏诚科技(北京)有限责任公司
  • dc.publisher
  • 扬州大学学报(人文社会科学版)
  • dc.publisher
  • Journal of Yangzhou University(Humanities and Social Sciences Edition)
  • dc.identifier.year
  • 2019
  • dc.identifier.issue
  • 06
  • dc.identifier.volume
  • v.23;No.138
  • dc.identifier.page
  • 89-96
  • dc.date.issued
  • 2019-11-29
  • dc.language.iso
  • 中文
  • dc.subject
  • 人工智能;司法知识库;知识图谱;事理图谱;证据链条;证据规则;风险预测
  • dc.subject
  • Artificial Intelligence;judicial knowledge base;knowledge graph;graph of facts and reasons;evidence chain;rule of evidence;risk prediction
  • dc.description.abstract
  • 司法与人工智能的融合是助力法院审判体系和审判能力智能化的基础。人工智能技术的飞速发展为构建司法知识库提供了必要的支持与帮助。当前的诉讼知识库是以树状层次化结构组织实体和节点关系,知识库中的实体关系主要表现为概念之间简单的"类属关系""同义词关系"。由于司法领域诸如证据链条、事理关系、法律判定规则等知识结构的复杂性,无法简单地组织、存储和应用数据结构和存储技术。是以,需要采用知识图谱和事理图谱相融合的技术,将司法领域中的专业术语和法律关系进行程序化表达,把标准化术语存储到证据要素知识模型中,以证据要素模板表达证据要素知识库,结合证据链条模型和证据规则模型的推理算法,从证据有效性、诉讼时效性、当事人法律行为规范性、诉讼请求合理性等方面对诉讼风险的提醒,以实现持久化的存储、事实推理和风险预测。
  • dc.description.abstract
  • The integration of judicature and artificial intelligence lays the foundation for the intellectualization of the court's judicial system and judicial capacity. The rapid development of artificial intelligence technology provides necessary support and assistance for the construction of judicial knowledge base. The current litigation knowledge base organizes the relationship between entities and nodes in a tree hierarchical structure. The entity relationship in the knowledge base is mainly manifested in the simple "generalization relationship" and "synonym relationship" among concepts. Due to the complexity of the knowledge structure in the judicial field, such as the chain of evidence, the relationship between facts and reasons, and the rules of legal decision, it is unable to organize, store, and apply data structures and storage technologies in a simple way. Therefore, it is necessary to adopt the technology of combining knowledge graph and graph of facts and reasons to procedurally express the professional terms and legal relations in the judicial field, storing standardized terms in the evidence element knowledge model. The knowledge base of strands of evidence is expressed by the template of strands of evidence. Combination of the reasoning algorithm of evidence chain model and evidence rule model, and reminder of litigation risk, from the aspects of the validity of evidence, the timeliness of litigation, the standardization of legal behavior of the parties, the rationality of litigation request and so on, help achieve the goal of persistent storage, factual reasoning and risk prediction.
  • dc.description.sponsorshipPCode
  • 2018YFC0830200;2018YFC0830202
  • dc.description.sponsorship
  • 国家重点研发计划“面向诉讼全流程的一体化便民服务技术及装备研究”项目(2018YFC0830200)中“研究面向多方证据关联分析的诉讼风险智能分析和结果预测技术”子课题(2018YFC0830202)
  • dc.identifier.CN
  • 32-1465/C
  • dc.identifier.issn
  • 1007-7030
  • dc.identifier.if
  • 0.260
  • dc.subject.discipline
  • G353.1;D926
回到顶部