From Bayesian Networks to Cognitive Science: How BayesiaLab Leads a Statistician to Hack Brains
Emmanuel Keita, sundiata.fr
At the heart of the of companies' information systems, close to decision-makers and operational staff, my professional experiences had led me to the observation that Business Intelligence (BI) did not provide a satisfactory answer to their real problems. What had changed in 20 years? Rainbow colors in the screens, OLAP cubes democratization, calculation speed… exciting — and also nightmare — for those in the IT department. But what is the real added value for the decision-makers and the workers? Curious about the potential of AI, I had just been trained in Python but, as a statistician, I saw nothing intellectually satisfying in my discovery of machine learning. Then a guy talked to me about "Bayesian networks."
That will be the starting point for this talk. An escapade into "augmented intelligence": Bayesian networks and BayesiaLab, encounters, change of relationship to the data, to beliefs, to the world and to myself… But, who am I, really ?
Emmanuel Keita has a multi-disciplinary background in mathematics and information system management:
- "Augmented Intelligence Evangelist" — he popularizes decision making discernment to different audiences: from high school to Google Zurich events, he focuses on human discernment in decision-making: not all is data!
- National defense auditor (France), Emmanuel was contributor of the "Villany report" — AI’s French strategy — (03/2018) and autor of an article "Le potentiel des réseaux bayésiens" in Défense & Stratégie Internationale (04/2019).
- Datamanager and statistician consultant in pharmaceutical industry (GSK), in charge of Statistics of Fujifilm, France.
- Head of the Business Intelligence division of Groupe Soufflet (agri-food industry).
- Emmanuel recently stopped Sundiata’s adventure for new challenges.