Start Date
10-12-2017 12:00 AM
Description
This paper designs and evaluates a customer emotion management system that aims to improve service operations of call centers by enabling role playing. The system is based on an array of data analytical techniques – ensemble support vector machine (E-SVM), convolutional neural network (CNN), recurrent neural network (RNN) – to automatically quantify customer emotions by analyzing both the acoustic and linguistic features of phone calls. Founded on the quantified and visualized real-time customer emotion, the system suggests effective service strategies for agents to guide customers. This acts as a useful tool for agents to manage their roles and for managers to monitor service operations. An empirical test of system performance illustrates that the emotion scores generated by the system can effectively indicate problematic service calls. Overall, the system highlights the benefit of IT on service encounter and showcases the business value of data analytical techniques in a real business context.
Recommended Citation
Guo, Yiting; Li, Yilin; Wei, Qiang; and Xu, Sean Xin, "IT-Enabled Role Playing in Service Encounter: Design a Customer Emotion Management System in Call Centers" (2017). ICIS 2017 Proceedings. 27.
https://aisel.aisnet.org/icis2017/DataScience/Presentations/27
IT-Enabled Role Playing in Service Encounter: Design a Customer Emotion Management System in Call Centers
This paper designs and evaluates a customer emotion management system that aims to improve service operations of call centers by enabling role playing. The system is based on an array of data analytical techniques – ensemble support vector machine (E-SVM), convolutional neural network (CNN), recurrent neural network (RNN) – to automatically quantify customer emotions by analyzing both the acoustic and linguistic features of phone calls. Founded on the quantified and visualized real-time customer emotion, the system suggests effective service strategies for agents to guide customers. This acts as a useful tool for agents to manage their roles and for managers to monitor service operations. An empirical test of system performance illustrates that the emotion scores generated by the system can effectively indicate problematic service calls. Overall, the system highlights the benefit of IT on service encounter and showcases the business value of data analytical techniques in a real business context.