Time to Transfer: Predicting and Evaluating Machine-Human Chatting Handoff
This work is significant for customer service operations and any domain utilizing chatbots, as it aims to improve user satisfaction by enabling timely human intervention when chatbots struggle.
This paper addresses the problem of when to transfer a chatbot conversation to a human agent, proposing a Difficulty-Assisted Matching Inference (DAMI) network to predict transferable utterances. They also introduce a new evaluation metric, Golden Transfer within Tolerance (GT-T), and collect two new datasets to validate their model.
Is chatbot able to completely replace the human agent? The short answer could be - "it depends...". For some challenging cases, e.g., dialogue's topical spectrum spreads beyond the training corpus coverage, the chatbot may malfunction and return unsatisfied utterances. This problem can be addressed by introducing the Machine-Human Chatting Handoff (MHCH), which enables human-algorithm collaboration. To detect the normal/transferable utterances, we propose a Difficulty-Assisted Matching Inference (DAMI) network, utilizing difficulty-assisted encoding to enhance the representations of utterances. Moreover, a matching inference mechanism is introduced to capture the contextual matching features. A new evaluation metric, Golden Transfer within Tolerance (GT-T), is proposed to assess the performance by considering the tolerance property of the MHCH. To provide insights into the task and validate the proposed model, we collect two new datasets. Extensive experimental results are presented and contrasted against a series of baseline models to demonstrate the efficacy of our model on MHCH.