Abstract:

Customer complaints in the telecommunications industry are increasing and urgently need to be handled efficiently. Based on the characteristics of telecommunication customer complaint data, this paper proposes an improved ensemble learning classification method for high-dimensional data. The proposed method comprehensively considers the text information and the customer communication status information in the customer complaint, and then based on the Random Subspace, using the Support Vector Machine（SVM） as the base classifier, and using the Evidential Reasoning（ER） rule as a new ensemble strategy to construct classification model to classify the telecom customer complaints. The proposed model and method are validated on the customer complaint data of a telecommunication company. The experimental results show that the proposed method can significantly improve the accuracy of customer complaint classification and the efficiency of complaint handling.

Key words: customer complaint classification, Random Subspace, Support Vector Machine（SVM）, evidential reasoning rule

摘要：

电信业的客户投诉不断增多而又亟待高效处理。针对电信客户投诉数据的特点，提出了一种面向高维数据的改进的集成学习分类方法。该方法综合考虑客户投诉中的文本信息及客户通讯状态信息，基于Random Subspace方法，以支持向量机（Support Vector Machine，SVM）为基分类器，采用证据推理（Evidential Reasoning，ER）规则为一种新的集成策略，构造分类模型对电信客户投诉进行分类。所提模型和方法在某电信公司客户投诉数据上进行了验证，实验结果显示该方法能够显著提高客户投诉分类的准确率和投诉处理效率。

关键词: 客户投诉分类, Random Subspace方法, 支持向量机, 证据推理规则