|
Data Science Colloquium Series 2 - 3 pm, NKN Hall Using Artificial Intelligence and Machine Learning to evaluate and improve Advertising: New solutions to Old problems Ashok Kalidas Kantar Analytics, Chennai. 30-08-19 Abstract With increasing computational power, widespread availability of data and development of powerful deep learning algorithms, the world of advertising evaluation is undergoing profound change. In particular, it is now possible to use advanced modeling techniques based on AI/ML to answer issues such as: * Is a particular advertisement likely to succeed in the market? * What type of emotion (if any) is a particular advertisement likely to evoke in a viewer? * Can we diagnose why some ads succeed and others fail? Can we use this to “optimize” elements of an ad? Elements of an ad will in general, include video (or images), audio, and text * How do different elements of an ad-campaign influence ROI, and how can we ensure that the “right” advertisement(s) are shown to the “right” audience with the “right” frequency * Do advertisements “travel” across countries? If an advertisement does well in a particular country, will it work well in a different one? In which one(s)? In this presentation, we go through some of Kantar’s latest and cutting edge work in this area. We provide an overview of our approach to answering some of the above questions, present examples of settings in which we have been successful and highlight some of the challenges in these inherently difficult problems. About the speaker: Dr Ashok Kalidas is Head of Data Science and Innovation at Kantar – a large multinational company that is a leader in marketing insights and analytics. Ashok provides data science based consulting services to Fortune 50 companies across a number of categories such as technology, retail, entertainment, healthcare, technology and financial services. In addition, Ashok leads an Innovation Lab which is responsible for evolving the next generation of Kantar’s products and services. In particular, a major focus of his effort is building AI/ML based solutions for analysis of video, images, and text data. Ashok holds a Ph.D. from the Engineering school at the Massachusetts Institute of Technology at the intersection of Statistics, Operations Research and Transportation, and an undergraduate degree from IIT Madras. He has worked with Kantar in various capacities in their US and India offices and is currently based in Chennai.
|
