Domain-Driven Data Mining: Examples and a Deep Learning Framework


Isn’t all data mining domain-driven? This talk will first briefly setup what “data mining” and “domain-driven” mean, and will then present several examples over the last two decades of work in the data mining that is domain-driven. From these examples, we will try to make the case that, while the domain always matters and plays a role, there are different extents to which this happens. We will also use examples to show the value of “thinking domain-driven”. Specifically, in addition to providing actionable intelligence and solving important problems in the domain, this perspective often also yields novel methods and insights that can then be applied more broadly (or “domain free”) as well. When this happens, “thinking domain-driven” can provide exponential benefits as well. We will use these discussions to present a general framework for domain-driven deep learning in business research and use this framework to show how researchers can position their own papers/ideas in a manner that makes it easier to highlight where significant contributions. Finally, while we have always viewed data as the source from which we need to learn, there is increasing understanding that the impact of the domain itself on the data can lead to unique situations that need newer methods to handle. We will use fairness in machine learning as a context to illustrate this dynamic and show how this leads to significant potential for new research.


Balaji Padmanabhan is the Anderson Professor of Global Management and Professor of Information Systems at USF’s Muma College of Business, where he is also the Director of the Center for Analytics and Creativity. He has a Bachelor’s degree in Computer Science from Indian Institute of Technology (IIT) Madras and a PhD from New York University (NYU)’s Stern School of Business. He has worked in data science, AI/machine learning and business analytics for over two decades in the areas of research, teaching, business engagement, mentoring graduate students and designing academic programs. During this time he has also worked with over twenty firms on machine learning and data science initiatives in a variety of sectors including financial services, technology, healthcare, manufacturing, services and non-profit. He has published extensively in data science and related areas at premier journals and conferences in the field and has served on the editorial board of leading journals including Management Science, MIS Quarterly, INFORMS Journal on Computing, Information Systems Research, Big Data, ACM Transactions on MIS and the Journal of Business Analytics.