The Center for Statistics and Applications in Forensic Evidence (CSAFE) wrapped up its fall webinar series and short courses in November. The webinar series, which included four sessions, provided an in-depth look at the latest forensic science research and resources available to the community.
New this fall was a three-part short course for forensic practitioners interested in learning the basics of machine learning. Machine Learning for Forensic Practitioners provided an overview of machine learning and its application to forensic evidence. The course covered supervised learning algorithms, including classification trees, random forests and neural networks.
CSAFE also offered Statistical Thinking for Forensic Practitioners. This introductory short course covered fundamental concepts from probability and statistics motivated by forensic issues. This course was expanded to four sessions this fall.
If you missed any of the live webcasts this fall or want to watch a webinar again, links to the recordings are available below.
The CSAFE webinar series will return in February with a new lineup of topics and speakers. Short courses will begin in March and will focus on probability and statistical concepts and their use in forensic science.
For the latest event announcements, follow @CSAFE_CoE on Twitter or visit forensicstats.org/events. To have event announcements delivered to your inbox, subscribe to the CSAFE newsletter at https://forensicstats.org/news-events/monthly-csafe-newsletters/.
The fall webinar series and short courses are sponsored by the National Institute of Standards and Technology (NIST) through cooperative agreement 70NANB20H019.
Fall 2022 Webinars
This webinar was held Aug. 25 and was presented by Danica Ommen, assistant professor of statistics at Iowa State University. The recording is available at https://forensicstats.org/blog/portfolio/ensemble-slrs-for-forensic-evidence-comparison/.
To strengthen the statistical foundations of forensic evidence interpretation, likelihood ratios and Bayes factors are advocated to quantify the value of evidence. Both methods rely on formulating a statistical model, which can be challenging for complex evidence. Machine learning score-based likelihood ratios have been proposed as an alternative in those cases. Under this framework, a (dis)similarity score and its distribution under alternative propositions are estimated using pairwise comparisons, but pairwise comparisons of all the evidential objects result in dependent scores. While machine learning methods may not require distributional assumptions, most assume independence. Ommen introduces a sampling and ensembling approach to remedy this lack of independence. She generates sets where assumptions are met to develop multiple score-based likelihood ratios later aggregated into a final score to quantify the value of evidence.
This webinar was held Sept. 15 and presented by Rachel Longjohn, a Ph.D. student in statistics at the University of California, Irvine (UCI), and Padhraic Smyth, Chancellor’s Professor in computer science and statistics at UCI. The recording is available at https://forensicstats.org/blog/portfolio/tutorial-on-likelihood-ratios-with-applications-in-digital-forensics/.
To date, digital forensics research has largely focused on extracting and reconstructing information from devices and the cloud. In comparison, there has been relatively little work on statistical methodologies that can be used to analyze such data after this step. In this webinar, the presenters will discuss statistical analyses in digital forensics, focusing on likelihood ratios and ideas from Bayesian statistics. There will be three parts to the webinar: introduction to likelihood ratios, the development of likelihood ratios in the context of digital forensics and applications to real-world datasets related to digital evidence.
This webinar was held Oct. 18 and presented by Alicia Carriquiry, director of CSAFE and Distinguished Professor and President’s Chair in Statistics at Iowa State University, and Stephanie Reinders, a CSAFE research scientist. The recording is available at https://forensicstats.org/blog/portfolio/handwriter-a-demonstration-and-update-on-csafe-handwriting-analysis/.
Forensic handwriting analysis relies on the principle of individuality: no two writers produce identical writing, and given enough quality and quantity of writing, it is possible to infer whether two documents were written by the same person. CSAFE is working on semi-automated methods suitable for closed or open sets of reference writers and for examining samples at the level of words or the level of graphical structures similar but not identical to graphemes. Carriquiry will describe each method in the webinar and show initial but promising results.
This webinar was held Nov. 21 and presented by Henry Maynard, chair of the American Society of Crime Laboratory Directors (ASCLD) Forensic Research Committee (FRC) and a board member on the ASCLD Board of Directors. The recording is available at https://forensicstats.org/blog/portfolio/the-ascld-frc-and-you-a-collaboration-worth-investigating/.
Over the last few years, the American Society of Crime Laboratory Directors (ASCLD) Forensic Research Committee (FRC) has created and launched tools to help advance forensic science research and further research collaborations within the community. From this presentation, participants will learn about forensic science research needs; the Laboratories and Educators Alliance Program (LEAP), which enables research partnerships; repositories for forensic research, evaluation, and validation efforts; executive research summaries; the research collaboration hub; and Lighting Talks. This presentation is for individuals looking to become more engaged in forensic science research and want to learn more about the opportunities the ASCLD FRC is creating to benefit the forensic science community.