Conference Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 393 - NLP and Text Analysis
Type: Contributed
Date/Time: Wednesday, August 10, 2022 : 8:30 AM to 10:20 AM
Sponsor: Section on Statistical Learning and Data Science
Abstract #320924
Title: Comparing Handwriting in Questioned Documents
Author(s): Alan Julian Izenman*
Companies: Temple University
Keywords: Handwriting comparisons; Historical background; Challenges; dynamic model; Legal systems; Error rates
Abstract:

In this talk, we discuss the current status of handwriting comparisons in the courts. We provide some historical background into the identification and comparison of handwriting specimens. We mention how the Roman, English, and United States legal systems learn from each other to treat comparisons of questioned documents. We discuss a number of unusual challenges in studying such documents. We describe how a handwriting examination is conducted and different statistical approaches to analyzing handwriting comparisons. In particular, we describe a dynamic model for handwriting that uses a second-order nonhomogeneous linear differential equation model to represent a handwriting sample. We present the results of published experiments that attempt to estimate error rates for handwriting comparisons by Forensic Document Examiners (FDEs) and non-FDEs.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2022 program