贝叶斯网络分类模型研究及其在信用评估中的应用

基于概率估计的贝叶斯及贝叶斯网络分类模型,拥有其它数据挖掘工具所不具备的优势。在分析贝叶斯及贝叶斯网络分类模型基础上,结合最小风险决策准则,提出了一种新的信用评估模型。在实际数据集上采用交叉验证方式进行了测试。实验结果表明基于最小风险决策准则的贝叶斯及贝叶斯网络分类模型可以有效地减少信用评估风险。

Study of Bayesian Network Classification Models and Its Application in Credit Scoring

Bayesian classifiers and Bayesian network classifiers are based on the probability estimate,which possesses some predominance than the other data mining tools. Via the analysis of the structure of Bayesian classifiers and Bayesian network classifiers,combining minimum overall risk rule,a new credit scoring model is proposed on the risk classification.They are tested by cross validation with a real data set according to minimum overall risk rule.The experimental results show that Bayesian classifiers and Bayesian network classifiers on minimum overall risk rule can decrease the risk of credit scoring effectively.