Intelligent auditing method and system based on knowledge graph anomaly reasoning

The invention relates to the technical field of intelligent auditing, in particular to an intelligent auditing method and system based on knowledge graph anomaly reasoning, and provides the following scheme: firstly, sorting typical auditing index items of different types of engineering projects in internal auditing, and forming a risk feature library according to project ranges and project contents; secondly, checking to-be-audited content according to an audit violation judgment strategy, obtaining a risk point set label of the to-be-audited content, constructing a knowledge reasoning neural network, taking the risk point set label of the to-be-audited content as an input parameter for training, obtaining risk similarity, and finally matching similar cases according to the risk similarity. And mapping to a knowledge graph to obtain the operation principle and suggestion of the exception. 本发明涉及智能审计技术领域，尤其涉及基于知识图谱异常推理的智能审计方法及系统，本发明提出以下方案，首先梳理内部审计中不同类型工程项目的典型审计指标项，根据的项目范围和项目内容，形成风险特征库，其次根据审计违章判定策略对待审计内容进行排查，获取