Since customer consumption attributes are multi-dimensional, related and uncertain, this paper proposed a customer consumption classification model based on rough set and neural network (RS-NN). Due to the rough set characteristics of customer consumption classification, the research framework of this paper was constructed by preprocessing knowledge space, establishing classification model and applying the classification model. Besides, based on RS, this paper also described consumption attributes reduction, classification rule extraction, and original topology of RS-NN construction, network model training and testing. Then a case study on telecom customers shows that RS-NN is better than BP-NN in construction, classification efficiency and prediction accuracy, which means RS-NN is an effective and practical new method for customer classification. © 2011 Springer-Verlag.
CITATION STYLE
Yinghong, W., Xiaopeng, C., Ying, Y., & Wanping, H. (2011). Research on the customer consumption classification model based on RS-NN. In Communications in Computer and Information Science (Vol. 224 CCIS, pp. 386–396). https://doi.org/10.1007/978-3-642-23214-5_51
Mendeley helps you to discover research relevant for your work.