Eduworks (United States)

Swofford & Champod (2022) FSI Synergy article 100220 reports the results of semi-structured interviews that asked interviewees their views on probabilistic evaluation of forensic evidence in general, and probabilistic evaluation of forensic evidence performed using computational algorithms in particular. The interview protocol included a leading question based on the premise that machine-learning methods used in forensic inference are not understandable even to those who develop those methods. We contend that this is a false premise.

Abstract Implementing competency-based training, education, and talent management requires models that compute and apply competencies to answer three questions: Does a person or team possess a given competency, how likely are they to successfully apply or demonstrate it, and what is the best way to acquire it? This paper argues that the underlying mathematical models should be treated as separate, although interrelated, and presents a framework for creating and computing such models that includes notions such as the level of a competency and conditions under which it is performed and that takes experience, practice, knowledge and skill decay, and the spacing effect into account. The source data for the computations outlined in this paper, which are described in the most detail for the model that ascribes possession of a competency, consists of assertions that in practice are derived from results reported by training systems. These computations reflect current practice in competency modeling and include “rollup rules” that respect relations among competencies. This paper includes an analysis of how these models appear and are used in intelligent tutoring systems.KeywordsCompetenciesAdaptive instructional systemsMathematical modelsExperiential trainingAssertionsState modelPredictive model