摘要
本研究主要找出影響信用卡消費金額因子，並建立預測模型。本研究使用迴歸分析進行因子篩選，接著使用迴歸分析與倒傳遞類神經進行預測，比較兩種方法對預測結果的優劣。觀察類神經網路預測模式與迴歸分析預測模式在信用卡消費金額的預測結果，類神經網路預測模式於信用卡消費金額有較佳的表現。
關鍵字：資料探勘、迴歸分析、類神經網路、參數設定。
ABSTRACT
The purpose of this research is to find out factors affecting the credit card spending and to build predicting models. With Regression Analysis, it was found the important factors. By using the factors found in the Regression Analysis as input, we built a Regression predicting model and a neural networks predicting model, and compare the results of two methods to make sure which method is the better one. After the predictions using the neural network models and the Regression predicting models at credit card spending were reviewed, neural network approach appeared to perform better at the credit card spending.
Keywords: Data Mining; Regression Analysis; Neural networks; Parameter settings
|