基于SMC-RS-LSSVM的电子商务客户流失预测模型

为提高个体层次上客户流失预测的精度,建立了基于SMC-粗糙集-最小二乘支持向量机的电子商务客户流失预测模型.该模型首先利用SMC模型计算出客户活跃度,以0.5为阈值判断出客户流失状态,识别出正判客户和错判客户;其次应用粗糙集理论约简出重要的客户流失预测指标体系,然后将训练样本送入最小二乘支持向量机进行学习和训练,进而对测试样本的客户流失状态进行判别.利用某网上商场的2525名客户样本进行电子商务客户流失预测实证研究,结果表明:与SMC模型、BP神经网络模型、最小二乘支持向量机模型相比,该模型对测试样本预测精度更高,是一种更为有效和实用的客户流失预测方法.

E-business customer churn prediction based on integration of SMC, rough sets and least square support vector machine

To improve individual customer churn prediction accuracy, integrated model of SMC, rough sets (RS) and least squares support vector machines (LSSVM), i.e., SMC-RS-LSSVM, for E-business customer churn prediction was proposed in this paper. Firstly, customers' active probabilities were obtained by using SMC model to identify customer churn status with the threshold of 0.5. The training and testing samples were formed by the correctly identified customers and incorrectly identified customers respectively. Then the attributes were reduced using rough sets and LSSVM was trained with training samples. Lastly, the trained LSSVM model was used to identify customer churn status of testing samples. Taking 2525 customers in an E-shop as samples, empirical results show that, compared with SMC, BP neural network and LSSVM models, integration model of SMC-RS-LSSVM is an efficient and practical tool for E-business customer churn prediction of testing samples.