Korean Customer Service Associate Assist System Based on Machine Learning

We propose a novel system that helps customer service associates answer customers' questions. The algorithm of the proposed system is as follows: First, when a customer asks a question to an associate, the question is transferred to a machine learning based assist system. Then, the system compares the received question to the questions in a predefined FAQ list. After that, the system outputs top five similar questions and the corresponding answers. Finally, the associate selects one question among five recommendations. As a result, additional training data are automatically generated by the selection of the associate, and they are used for re-training of the system. The experimental result shows that the automatically generated data of the proposed system recursively reinforces the accuracy of the system. KeywordsMachine learning, Customer service, FAQ bot, BiGRU, Natural language processing