首页|基于知识图谱与加权贝叶斯机制的刑侦推断模型

基于知识图谱与加权贝叶斯机制的刑侦推断模型

扫码查看
随着信息化和数字化的高速发展,各类人工智能方法在现代公共安全领域得到充分应用,其中知识图谱以其强大的知识表征能力和数学建模能力,大大提高了智能决策的可解释性,成为领域内研究和应用的热点.一直以来,基于贝叶斯机制的推断模型是公安情报与刑侦分析的重要手段.本文将公安知识图谱引入传统贝叶斯模型,在充分考虑参数耦合性的基础上,对后验概率进行加权更新,有效提高了推断的准确性.本文回避敏感的真实数据,以《红楼梦》中"绣春囊"一案为例完成实证分析,科学地揭秘了经典悬案的真相.本文提出的基于知识图谱的自适应加权方案有效提高了贝叶斯模型的效率,是人工智能助力智慧公安的一种有益尝试.
Criminal Investigation Inference Model Based on Knowledge Graph and Weighted Bayesian Mechanism
With the rapid development of informatization and digitization,various artificial intelligence methods have been extensively applied in the field of modern public security.Among them,knowledge graphs,with their powerful knowledge representation and mathematical modeling capabilities,have greatly improved the interpretability of intelligent decision-making,making them a hot topic of research and application in this field.Bayesian inference models based on the Bayesian mechanism have always been important means for public security intelligence and criminal investigation analysis.In this paper,we introduce the knowledge graph of public security into the traditional Bayesian model.By fully considering the coupling of parameters,we effectively update the posterior probability with weighted updates,significantly improving the accuracy of inference.Taking the case of"Embroidered Spring Bag"in the novel"The Story of the Stone"as an empirical analysis,this paper scientifically reveals the truth of a classic case.The proposed scheme of using knowledge graph's search adaptive weighting effectively improves the efficiency of the Bayesian model,making it a beneficial attempt to empower intelligent public security with artificial intelligence.

knowledge graphBayesian modelrelationship weightsadaptive weighting

李波、章勇、胡誉骞

展开 >

华中师范大学 数学与统计学学院,湖北 武汉 430079

合肥第六中学,安徽 合肥 230061

知识图谱 贝叶斯模型 关系网络权重 自适应加权

国家自然科学基金面上项目湖北省高校省级教学研究项目武汉市科技计划知识创新专项项目

6237701920220832022010801010273

2024

数学建模及其应用

数学建模及其应用

影响因子:0.215
ISSN:
年,卷(期):2024.13(3)
  • 9