A practical approach to dialogue response generation in closed domains
This work addresses the problem of automating customer service responses for companies like Amazon, but it is incremental as it builds on existing template-based and similarity methods.
The authors tackled dialogue response generation in customer service by developing a weakly supervised dual encoder model that selects answer templates from past agent responses, achieving over 70% coverage of customer inquiries and significantly higher relevance than a tf-idf baseline.
We describe a prototype dialogue response generation model for the customer service domain at Amazon. The model, which is trained in a weakly supervised fashion, measures the similarity between customer questions and agent answers using a dual encoder network, a Siamese-like neural network architecture. Answer templates are extracted from embeddings derived from past agent answers, without turn-by-turn annotations. Responses to customer inquiries are generated by selecting the best template from the final set of templates. We show that, in a closed domain like customer service, the selected templates cover $>$70\% of past customer inquiries. Furthermore, the relevance of the model-selected templates is significantly higher than templates selected by a standard tf-idf baseline.