The Daily Bulletin

Please join us for “Emerging Aspects of Data Science,” a presentation by Ernest Fokoue, professor in the School of Mathematics and Statistics at the Rochester Institute of Technology, on Tuesday, May 7, from 5:55 to 7:05 p.m. in Science and Mathematics Complex 151. This talk is part of the Data Science and Analytics Seminar Spring 2024 Series.

Talking Points

• Classical approaches to statistical analysis
• State-of-the-art methods of predictive modeling
• Inevitability of the need for holistic approaches to data science
• Unifying power of the non-pinnable nature of data 

Abstract
In the practical exploration of potential patterns underlying data, it does not take long for one to realize that simple conclusions are rarely satisfactory. As it were, data—real-life data—does not lend itself to the kind of reductionist summarization often sought and desired by traditional mathematical scientists. More often than not, there are many facets to the kind of patterns that any dataset might exhibit. This echoes something akin to the passage from classical Newtonian physics to Einsteinian relativistic physics when one has to admit the limitations of the assumptions of a fixed reference and welcome a new and richer paradigm.

This talk is not as philosophical as the earlier sentences might suggest, but perhaps conversational and intended to drive home the crucial importance of resorting to a more flexible and more holistic view of modeling and data analytics, one that is no longer uncomfortable for the failure of simplistic and reductionistic summarization, as one needs to embrace the beauty, richness, and power of subjectivity in data science! 

About Ernest Fokoue
Ernest Fokoue earned his Ph.D. in statistics at the University of Glasgow in the United Kingdom. He was a postdoctoral research fellow at the Statistical and Applied Mathematical Sciences Institute (SAMSI). Before joining the Rochester Institute of Technology, he held faculty positions at the Ohio State University and Kettering University. He is the coauthor of the Springer graduate textbook Principles and Theory for Data Mining and Machine Learning and has been intensively and extensively involved in the interface between statistical science and artificial intelligence. He is a strong advocate of a complete approach to statistical machine learning and data science comprising a non-degenerate coverage of applications, computation, methodology, and theory (ACMT), the staple of his research and teaching. A passionate lover of mathematical sciences from his earliest childhood, Dr. Fokoue often summarizes his philosophy as follows: “While the letter of mathematics provides great delight to the seeker/researcher/practitioner, the spirit of this mother of all disciplines bestows experiences of a transcendental, nay ineffable nature. Therefore, seek the latter not the former.”