Interactive AI – Keeping Users in the Loop
With so much of information around, it has become utmost important for research communities to find answers related to enabling easier information access and easier information consumption for end-users. Recent advances in user-centric design methodologies focus on technology-enabler (AI-based) to help users to concentrate on the information rather than how to procure and curate them. To design such a echnology enabler research communities have to answer few related research questions. They are: what to deliver (content), whom to deliver (user’s context), how to deliver (delivery method and channel) and when to deliver (user’s context). Most of the state-of-the-art in the recommendation system or in knowledge management have addressed problems related to answering the “what” part of the overall problem; when, how, whom parts are still remained open to further research efforts. In my research, I have been trying to find answers for these open research questions. In my Ph. D. thesis, I have demonstrated how
by analysing and modelling the context and the user-interactions in a system, we can efficiently find out the answers for the research questions related to “how”, “when” and “whom” parts. In my current work in Fractal Analytics, as a follow-up on these research efforts, I have been working on designing a personalized business analytics delivery platform for business users in enterprises. In this talk, I would discuss about few of these research efforts in a nutshell.
About the Speaker
Soudip Roy Chowdhury is currently working in the capacity of Director of Data Science Fractal Products, Fractal Analytics. In his current role he is responsible for designing immersive data analytics products for business users in enterprises. Prior to Fractal, Soudip worked for 12 years in several research institutes both in academia (Inria, Paris, University of Trento, Italy) and in industry (IBM india Research lab, Rakuten Institute of Technology, Paris etc.) in India and abroad. His Ph.D. topic was on assisted reuse of pattern-based composition knowledge for mashup development. Since his Ph.D, he has been working on topics related to knowledge management, recommendation system, data analytics using data-mining and machine learning algorithms.