基于SVM机器学习技术的企业智能化审计建模优化  

Optimization of enterprises intelligent audit modeling based on SVM machine learning technology

作　　者：蔡玲嘉 CAI Lingjia(Audit Center of Guangdong Power Grid Co.,LTD.,Guangzhou 510060,China)

机构地区：[1]广东电网有限责任公司审计中心,广东广州510060

出　　处：《粘接》2023年第5期139-142,共4页Adhesion

摘　　要：机器学习是人工智能的核心,将其应用于企业审计中,提升企业审计智能化水平。研究从用户、内部业务流程、学习和成长、财务4个角度构建了审计智能化评价指标,并采用经典机器学习算法支持向量机建立企业智能化审计评价模型。为提升支持向量机模型性能,采用回溯搜索优化算法对支持向量机核函数进行优化,将构建的模型与GA-SVM、PSO-SVM进行对比。结果表明:BSA-SVM模型的分类识别准确率最高为94.5%,同时迭代时间最短为36.28 s。Machine learning is the core of artificial intelligence,and it can be applied to the audit of enterprises to improve the level of enterprise audit intelligence.In this paper,intelligent audit evaluation indicators were con⁃structed from four perspectives:users,internal business processes,learning and growth,and finance,perspec⁃tives of users,internal business processes,learning and growth,and finance,then classical machine learning algo⁃rithm was used to support vector machines to establish an intelligent audit evaluation model for power enterprises.In order to improve the performance of the support vector machine model,a backtracking search optimization algo⁃rithm was used to optimize the kernel function of the support vector machine,and the constructed model was com⁃pared with GA-SVM and PSO-SVM.Results showed that the classification recognition accuracy of the BSA-SVM model is the highest 94.5%,and the shortest iteration time is 36.28.

关 键 词：支持向量机 回溯搜索优化算法 智能化审计 技术 建模 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]