Multivariate data extraction method of enterprise digital internal audit based on knowledge map and LDA model

The explosive growth of enterprise data, with diverse data types and complex structure, provides abundant information resources for enterprise's operational decision-making, risk management and compliance review. However, how to extract valuable information from these massive data efficiently and accurately has become a key problem to be solved urgently in the digital internal audit of enterprises. Therefore, this paper proposes a multivariate data extraction method for enterprise digital internal audit based on knowledge map and LDA model. Define ontology and entity, construct multivariate data security knowledge map of enterprise digital internal audit, obtain the thematic correlation between documents and words, and construct LDA model; The features are aggregated to get the entity data vector, and the multivariate data extraction of enterprise digital internal audit is realized. The experimental results show that the information coverage of this method is high, and it shows significant advantages in the cost of multivariate data extraction in enterprise digital internal audit.