摘要
笔者以葡萄牙商业银行2008年到2013年银行客户数据集为例,使用数据挖掘技术预测银行客户接受银行长期存款情况。通过决策树、类神经网络(ANN)和逻辑特回归(logistic)等技术构建五个预测模型,通过混淆矩阵结果看出,ANN模型预测效果最高,其预测精度达92.4%。通过预测模型为银行决策层提供决策依据,减少银行成本投入,提高了银行营销人员的促销成功率。
The author takes the example of bank customer data set from 2008 to 2013 in Portugal Commercial Bank.Using data mining technology to predict customers'deposits whether or not to accept long-term.Using the decision tree,ANN and logistic to prediction model,the results of confusion matrix show that the ANN model has the highest prediction accuracy,and the prediction accuracy is 92.4%.Through the prediction model to provide decision-making for the bank decision-making level,reduce bank cost inputs,and improve the bank marketing personnel's promotion success rate.
作者
黄晓茜
Huang Xiaoqian(Tongji University,Shanghai 200000,China)
出处
《信息与电脑》
2017年第9期133-135,140,共4页
Information & Computer