Augmenting Customer Support with an NLP-based Receptionist
This work addresses customer support automation for a real-estate company, but it is incremental as it applies existing methods (BERT and finite state machines) to a specific domain and dataset.
The paper tackled the problem of predicting client contact motivation for a real-estate company by deploying a conversational AI system using a Portuguese BERT model combined with structured data and a finite state machine, achieving human-level results on a dataset with 235 unbalanced labels and showing business benefits compared to classical NLP methods.
In this paper, we show how a Portuguese BERT model can be combined with structured data in order to deploy a chatbot based on a finite state machine to create a conversational AI system that helps a real-estate company to predict its client's contact motivation. The model achieves human level results in a dataset that contains 235 unbalanced labels. Then, we also show its benefits considering the business impact comparing it against classical NLP methods.