Lionel Yelibi

Lionel Yelibi

Applied Scientist & Quantitative Researcher - Emerging Capabilities Research Group

Discover Financial Services

UCT Statistical Finance

Biography

Welcome to my academic page! I am Lionel Yelibi, a data science researcher specializing in clustering methods for high-dimensional and time-series data. My expertise lies in developing algorithms that leverage complex systems and spin models, originally from physics, to analyze intricate datasets, including financial market data. I am particularly interested in creating efficient, scalable clustering methods, such as the Agglomerative Fast Super-Paramagnetic Clustering (ALC) technique, designed to identify and map hidden structures within interdependent data points.

My professional interests are at the intersection of computational modeling and real-world data applications, especially in finance, where clustering helps to detect relationships among stock behaviors and market dynamics. In my academic research, I explore how advanced clustering algorithms can improve pattern recognition and data interpretation in environments with high connectivity and complexity, striving to push the boundaries of what data science can reveal.

Interests
  • Econophysics
  • Quantitative Finance
  • Statistical Mechanics
  • Network Science
  • Machine Learning
  • Manifold Learning
Education
  • Ph.D. in Statistical Sciences (2028)

    University of Cape Town

  • M.Sc. in Statistical Sciences (2019)

    University of Cape Town

  • BSc in Physics

    Indiana University

Experience

 
 
 
 
 
Lead Research Scientist
Discover Financial Services
May 2022 – Present Chicago
 
 
 
 
 
Investment Research Analyst
Mass PRIM
Nov 2020 – Mar 2022 Massachussets
 
 
 
 
 
Machine Learning Researcher
Boston Fusion Corp.
Jan 2020 – Nov 2020 Massachussets
 
 
 
 
 
Research Assistant
University of Cape Town
Aug 2017 – Dec 2019 South Africa