Robustness Challenge with Dialog Systems, Myths and Solutions

Dialog systems (aka chatbots) are now key elements of digital channels in finance and banking environment. Dialog systems allows cost reduction, can increase customer satisfaction, and reduces errors. However, dialog technologies do not easily meet customer expectations. Machine to human conversation is a still a difficult aspect to handle, and recent advance in intent classification or dialog utterance handling using deep neural networks did not solve the question of robustness of dialog systems when deployed in production environment, with real human language interactions. In this communication, we will present innovative paraphrasing algorithms and dialog model generators we developed at National Bank of Canada to tackle this problem and improve robustness of dialog systems.

Éric Charton, Senior AI Director at National Bank of Canada

Eric Charton hold a Master in machine learning applied to voice recognition, and a Ph.D. in machine learning applied to Information extraction and natural language generation. He worked as scientist and research project coordinator in academic context in Europe (University of Avignon) and North America (CRIM, École Polytechnique de Montréal) before becoming head of search engine research and development at Yellow Pages Canada. Since March 2018, he is Lead AI Expert at National Bank of Canada.

Cookies help us deliver our services. By using our services, you agree to our use of cookies. Learn more