Evaluating Empathetic Chatbots in Customer Service Settings
This research addresses the problem of improving chatbot responses in customer service by making them more human-like and empathetic, benefiting both customers and companies.
The paper investigates the impact of training chatbots to recognize emotion and exhibit empathy in customer service settings. It found that a blended skills chatbot model trained with these capabilities is more likely to resemble actual human agent responses compared to a model without such training.
Customer service is a setting that calls for empathy in live human agent responses. Recent advances have demonstrated how open-domain chatbots can be trained to demonstrate empathy when responding to live human utterances. We show that a blended skills chatbot model that responds to customer queries is more likely to resemble actual human agent response if it is trained to recognize emotion and exhibit appropriate empathy, than a model without such training. For our analysis, we leverage a Twitter customer service dataset containing several million customer<->agent dialog examples in customer service contexts from 20 well-known brands.