摘要
高校管理者采用传统财务报表和历史经验无法满足高校财务管理信息化的需要。基于改进决策树算法组建高校财务管理和决策分析系统。首先组建高校财务管理和决策样本数据库,采用水库抽样算法均匀抽取高校财务管理和决策样本,在此基础上利用决策树算法将样本训练集定义为根节点,并将其划分为不同类型的子集,采用MBDT对决策树进行优化,有效提升了决策树的分类精度,且能给出各个数据样本的信息增益率。由此组建了能实现高校财务管理信息化的财务预警模型。实验仿真证明,提出的改进决策树算法具有较高的预测精度,在建设高校财务管理决策体系中具有重要的应用价值。
University administrators use traditional financial statements and historical experience cannot meet the needs of university financial management informatization.This paper builds the financial management and decision analysis system of university based on the improved decision tree algorithm.Firstly,a sample database established for university financial management and decision-making,and use a reservoir sampling algorithm to uniformly sample university financial management and decisionmaking samples.On this basis,use the decision tree algorithm to define the sample training set as the root node and divide it into for different types of subsets,MBDT is used to optimize the decision tree,which effectively improves the classification accuracy of the decision tree and can give the information gain rate of each data sample.Therefore,a financial early warning model which can realize the informatization of financial management in university is established.Experimental simulation proves that the improved decision tree algorithm proposed in this paper has high prediction accuracy and has important application value in the construction of university financial management decision-making system.
作者
蔡晶
CAI Jing(Finance Office,Huaqiao University,Quanzhou Fujian 3620000,China)
出处
《佳木斯大学学报（自然科学版）》
CAS
2022年第2期133-135,158,共4页
Journal of Jiamusi University：Natural Science Edition
关键词
改进决策树算法
高校财务管理
决策分析
仿真研究
improved decision tree algorithm
university financial management
decision analysis
simulation research
