We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Survey of Computerized Polygraph Scoring Algorithms Using Kircher Features.
- Authors
Nelson, Raymond
- Abstract
Published literature was surveyed for automated statistical classifiers that make use of a common set of physiological response features. These features were first introduced and described by researchers at the University of Utah during the 1980s, and have subsequently come to be described with the polygraph profession as Kircher features. These include the amplitude of increase for electrodermal and cardiovascular activity, along with a reduction of respiration activity. Constriction or reduction of vasomotor pulse amplitude can also be included. An interesting characteristic of these features, in addition to their statistical correlation with deception and truth-telling, they can be extracted from recorded time-series data both visually and via automated computer methods. Statistical classifiers based on these features include the following: Probability Analysis, a Rank Order Scoring System, an Objective Permutation method, a Bootstrap analysis method, the Empirical Scoring System/Multinomial, the Objective Scoring System (version 1 and 2), and the Objective Scoring System - version 3. Design characteristics of these analysis methods are summarized in the appendices.
- Subjects
LIE detectors &; detection; COMPUTER algorithms; UNIVERSITY of Utah
- Publication
Polygraph & Credibility Assessment, 2022, Vol 51, Issue 2, p74
- Publication type
Article