Forensic Data Science
Laboratory


Home



About Us

The Forensic Data Science Laboratory conducts research to develop methods for evaluation of forensic evidence that are based on relevant data, quantitative measurements, and statistical models; methods that:

  1. are transparent and reproducible;
  2. are intrinsically resistant to cognitive bias;
  3. use the logically correct framework for interpretation of evidence (the likelihood-ratio framework); and
  4. are empirically validated under casework conditions.

In order to develop methods that provide solutions for real forensic-evaluation problems, solutions that have a high probability of actually being adopted in casework, members of the Laboratory conduct research in collaboration with researchers and practitioners who have expertise in particular branches of forensic science.

Members of the Laboratory also conduct research on calibration and validation of forensic-evaluation systems, and on communication of forensic science to courts, research whose results are applicable across many branches of forensic science.

In addition to research, members of the Laboratory provide training in forensic inference and statistics to forensic practitioners and to lawyers, and contribute to the development of standards and guidelines for forensic science.



Selected Publications



Projects

Calibration and Validation of Forensic-Evaluation Systems

  • methods for calculation of likelihood ratios
  • empirical calibration of forensic-comparison systems
  • empirical validation of forensic-comparison systems

E3 Forensic-Speech-Science System (E3FS3)

  • standards, guidelines, and standard operating procedures
  • data-collection protocols and software
  • open-code software tools based on state-of-the-art automatic-speaker-recognition technology
  • library of validation reports
  • training for practitioners

Firearms

  • comparison of fired cartridge cases
  • data collection
  • development and validation of feature-based methods for calculation of likelihood ratios

Forensic Anthropology

  • development and validation of methods for calculation of likelihood ratios

Cell-Site Analysis

  • data collection
  • development and validation of methods for calculation of likelihood ratios

Fingerprints

  • data collection
  • development and validation of forensic-data-science methods for comparison of fingerprints and fingermarks
  • training for practitioners application for research funding under development

Communicating Forensic Science

  • understanding of likelihood ratios by legal-decision makers
  • understanding of validation by legal-decision makers

Speaker Recognition / Speaker Identification by Human Listeners

  • speaker recognition by earwitnesses
  • speaker identification by judges and juries

– Authorship Attribution

  • development and validation of methods for calculation of likelihood ratios

– Training

  • training in forensic data science, including:
    • the likelihood-ratio framework
    • calibration and validation
    • application to particular branches of forensic science
  • for:
    • forensic practitioners
    • lawyers
    • judges




Events

– Reading group

  • We have a weekly meeting at which we discuss published papers related to forensic data science. These are usually recently published papers. Sometimes we invite the authors of the papers to join us for the discussion. We usually meet online on Mondays at 09:00 UK time (for the last meeting of Jan 2022 through to the end of March 2022 we have moved to Fridays).

  • Meetings are open to Laboratory members, adjunct members, and invited guests only. If you have an existing relationship with the Laboratory, or one of more of its members or adjunct members, and would like to receive invitations to join the reading-group meetings, please send a request to the Laboratory Director. Requests will be considered on a case by case basis.

  • Schedule of past and upcoming readings

    • This schedule is updated as needed – please check here for the latest version.

    • Meeting participants: Please send suggestions for readings to Dr Basu.


European Academy of Forensic Science Conference


Laboratory Members


Geoffrey Stewart Morrison

  • Associate Professor of Forensic Speech Science, Aston Institute for Forensic Linguistics, Aston University
  • Director, Forensic Data Science Laboratory, Aston University


Phil Weber

  • Lecturer, Computer Science, Aston University


Nabanita Basu

  • Research Associate, Forensic Data Science Laboratory, Aston University


Patrick Geoghegan

  • Senior Lecturer, Biomechanical Engineering, Aston University

Former Member


Roberto Puch-Solis

  • Principal Investigator, Leverhulme Research Centre for Forensic Science, University of Dundee

Adjunct Members


Rachel Bolton-King

  • Associate Professor, School of Justice, Security and Sustainability, Staffordshire University
  • Visiting Researcher, Aston University


Ewald Enzinger

  • Senior Research Engineer, Eduworks Corporation
  • Visiting Researcher, Aston University


Rolf J F Ypma

  • Principal Scientist, Netherlands Forensic Institute
  • Visiting Researcher, Aston University


Cuiling Zhang

  • Professor and Vice Dean of School of Criminal Investigation, Southwest University of Political Science and Law
  • Visiting Researcher, Aston University


Claudia Rosas-Aguilar

  • Associate Professor, Instituto de Lingüística y Literatura, Universidad Austral de Chile
  • Visiting Researcher, Aston University

Advisory Board Member


Didier Meuwly

  • Principal Scientist, Netherlands Forensic Institute
  • Professor, Department of Data Management & Biometrics, University of Twente

Collaborators and partner organizations on specific projects are listed on project pages



Forensic Talks Interview with Geoffrey Stewart Morrison

Forensics Talks: Forensic Data Science

Recording of interview originally live-streamed 2021-04-08

https://www.youtube.com/watch?v=ysGEfPxTY-Q




Funding





http://forensic-data-science.net/

This webpage is maintained by Geoffrey Stewart Morrison and was last updated 2022-01-22