Luis Carvalho, PhD, will give the second talk in The USM Data Science Ensemble, a seminar series focused on the intersection of data science and real-world applications. We invite you to join us for this in-depth look at a practical application of data science in the real world, by joining this moderated Zoom link.
In Dr. Carvalho's talk, "Modeling Residential Burglaries in Boston: a Data Science Case," he will explain a new methodology for modeling occurrences of events in a network and use residential burglaries in Boston as a case study (a joint work with Liz Upton). Dr. Carvalho will discuss the core of their model: a functional network regression on node attributes using a Laplacian operator based on edge similarities as regularizer. He will show how usual regularization penalties can be cast as prior distributions on regression coefficients under a Bayesian setup, and propose a computationally efficient EM fitting procedure. For the case study, he will discuss how publicly available data from the city of Boston can be cleaned, visualized, and explored to give valuable insights into model building, calibration, and presentation and interpretation of results.
Luis Carvalho is an Associate Professor in the Department of Mathematics & Statistics at Boston University and director of the MSc in Statistical Practice program. Prior to joining BU ten years ago, he did his PhD in Applied Mathematics at Brown University investigating Bayesian approaches to high dimensional discrete inference. His main research interests are Bayesian and computational statistics and data science applications in many fields such as engineering, genetics, and social sciences.