This month we are pleased to welcome Nicholas Petraco to discuss the application of statistical computing to crime scene forensics. Yeah, that's right, we're getting all CSI in February!
The practice of qualitative physical evidence comparison in forensic science has come under great scrutiny since the 2009 publication of the NAS report on “Strengthening Forensic Science in the United States”. From a computational point of view, “evidence” from a crime is simply data. Modern machine learning techniques can make associations between data samples from a suspect, and evidence from a crime scene quantitative and objective. Specifically, given multivariate data constituting evidence in a crime, how likely is it that a conclusion of association with a suspect is erroneous?
This presentation will discuss how our group of forensic researchers and practitioners are using R to be able to answer this question. Projects discussed will be on the use of principal component analysis (PCA), canonical variate analysis (CVA), partial least squares discriminant analysis (PLS-DA), support vector machines (SVM), resampling, conformal prediction theory (CPT) and empirical Bayes methods to establish identification systems and estimated error rates on firearms and tool mark evidence, dust discrimination, fire debris evidence, footwear associations and questioned photocopier analysis.
Speaker Bio: Nicholas D. K. Petraco earned a bachelors degree in Chemistry from Colgate University in 1998 and a doctorate in Quantum Chemistry at University of Georgia, Center for Computational Quantum Chemistry in 2002. He was a postdoctoral fellow in Applied Mathematics at the University of Waterloo from[masked] where after he was appointed to an Assistant Professorship at John Jay College of Criminal Justice and The Graduate Center, City University of New York. Currently Nick is an Associate Professor at CUNY and belongs to the American Academy of Forensic Sciences (AAFS), the North Eastern Association of Forensic Scientists (NEAFS), the Society for Industrial and Applied Mathematics (SIAM) and the Institute for Electrical and Electronics Engineers (IEEE). His research interests are in the application of computing and advanced statistical pattern recognition methods to physical evidence.
Schedule will proceed as usual: food and networking starting at 6:15pm, Nicholas will begin promptly at 7pm, and we will migrate to a local watering hole afterwards.