针对纺织行业客户流失问题建立了基于支持向量机的预测模型。基于该行业预测客户流失指标属性多、相关系数高的特点,首先采用主成分分析法从多指标属性中筛选出客户流失的主要因素,有效地降低了支持向量机的训练维度。通过实际纺织行业的客户数据集测试,与普通支持向量机及其他传统预测模型进行比较,验证该模型具有良好的推广能力以及更高的精确性。
To deal with customer churn problem in textile industry, this paper set up prediction model based on support vector machine(SVM). Due to easily-correlated,multi-index of indicative attributes in churn data, adopted principal component analysis(PCA) to screen out the main factors from a great deal of indicative attributes in order to reduce the training dimension of SVM effectively. With the application and verification in real textile data set, the result demonstrates that this model has a better universal property with higher precision than others.