Société Générale

Chatbot to assist client advisors – exploiting unstructured data sources

Banking & Insurance


• The client advisors in banks cannot answer their clients questions if they require an intimate knowledge of the internal banking rules or of the legal framework.
• The experts these advisors rely on are particularly busy, so they cannot answer in less than two days
• We are provided with a raw database of questions asked by the advisors to the experts


• Prove the viability of a chatbot designed to answer complex banking questions
• Exploit an unstructured text-based data source without input from business


• Manual classification of a sample from the questions database (focus on real estate loans)
• Text preprocessing : normalization and ad hoc features
• Machine learning algorithm to classify questions into intents (questions with the same answer)
• Training a chatbot on this newly structured database, thanks to Rasa, an open-source framework
• User interface for a demonstrator
• Chabot for demonstration purposes, able to identify questions about 40 different complex topics
• Chatbot trained on a set of 2500 questions

Want to talk to us about your project ?
Contact us
Contact us

122 avenue du Général Leclerc
92100 Boulogne Billancourt

+33 (0)1 46 44 24 66

Join us

Come and play !
A strong team is where it all begins.

Join us