An Analysis of Unreliability of Competence Information in Learning Networks and the First Exploration of a Possible Technical Solution

Abstract

Automated competence tracking and management is crucial for an effective and efficient life-long competence development in learning networks. However, currently there is no systematic method to represent, measure, and interpret competence. In this paper, we analyze the problem of unreliability of competence information in learning networks. In tracking the development of competences in learning networks, a large amount of competence information can be gathered from diverse sources and diverse types of sources, which is subject to uncertainty and unreliable. This paper investigates information fusion technologies that may be applied to address the problem and that show promise as candidate solutions for achieving an improved estimate of competences by fusing (possibly inconsistent) information coming from multiple sources. This paper is intended to motivate educational technology researchers to learn more about information fusion, to perform studies with real and simulated data sets, and to apply in learning networks that may benefit from information fusion technologies.