By Ken Duffy (Northeastern University)

Talk Abstract: When you read that someone’s DNA was found at a crime scene, it can seem to be irrefutable evidence of their presence, yet miscarriages of justice occur based on that premise. Many of those have arisen due to shortcomings in past data analytic interpretation frameworks rather than resulting from issues with sample preparation.

The goal of this introductory talk is to explain how people are associated with a crime scene, how the complexities of crime scene samples result in interpretative challenges, how those difficulties are treated with current probabilistic frameworks, and how the output of the assessment is reported to juries through a verbal quantization of a log-likelihood ratio.

We will introduce a large scale dataset, the PROVEDIt database (https://lftdi.camden.rutgers.edu/provedit/), that has been made public to enable algorithmically minded experts to develop novel analytic methodologies and test them at scale based on realistic data. We will also describe recent developments that may expand the remit of DNA forensics to more complex samples, the new challenges they bring, and the work we have been doing to resolve them.

This talk is based on ongoing work with Catherine Grgicak (Rutgers) and Desmond Lun (Rutgers) as well as their teams.

Speaker Bio:From 2023, Ken R. Duffy is a professor at Northeastern University with a joint appointment in the Department of Electrical & Computer Engineering and the Department of Mathematics. He was previously a professor at Maynooth University in Ireland,where he was the Director of the Hamilton Institute, an interdisciplinary research centre with 40 affiliated faculty, and one of three co-Directors of the Science Foundation Ireland Centre for Research Training in Foundations of Data Science, which funds more than 100 PhD students. He is a co-founder of the Royal Statistical Society’s Applied Probability Section (2011), co-authored a cover article of Trends in Cell Biology (2012), is a winner of a best paper award at the IEEE International Conference on Communications (2015), the best paper award from IEEE Transactions on Network Science and Engineering (2019), and the best research demo award from COMSNETS (2022).