DIGITAL LIBRARY
ARTIFICIAL INTELLIGENCE APPLICATION TECHNIQUES FOR DEEP ASSESSMENT OF MULTIPLE CHOICE EDUCATIONAL SYSTEMS
1 WebDBTech (UNITED STATES)
2 Alexandria University (EGYPT)
About this paper:
Appears in: EDULEARN20 Proceedings
Publication year: 2020
Pages: 26-30
ISBN: 978-84-09-17979-4
ISSN: 2340-1117
doi: 10.21125/edulearn.2020.0030
Conference name: 12th International Conference on Education and New Learning Technologies
Dates: 6-7 July, 2020
Location: Online Conference
Abstract:
Learning is the quintessential differentiating aspect that favors humans on this planet. The ability to learn the accumulated knowledge from one human to the other and furthermore to prissiest this knowledge throughout the ages have been at the canter of all human civilizations. Education developed as a systematic way to institutionalize the learning process and scale it. Any evolving system needs to be able to adjust and adapt to improve over time. The main way by which any control system adjusts is to have a feedback loop, a way to assess and measure the current status in order to adapt. This process translates to assessments and testing in education, however, this process requires considerable manual effort. In recent years, especially with the advancement of electronic testing, multiple-choice assessments became prevalent as it can be automatically evaluated removing the need for human intervention. The main limitation of this method is the dichotomy of the result in contrast with a fully qualified answer. The conundrum is that the fully qualified answer requires an order of magnitude effort to grade, which is mostly manual and not easy to scale. In this paper, a novel approach is presented in which Artificial Intelligence is implemented in order to demonstrate how a multiple-choice assessment can be constructed so that a deeper analysis of the answers is attained. The paper further asserts the potential to render partial grades resulting in a superior assessment of the learner’s understanding.
Keywords:
Artificial Intelligence, Machine Learning, Education, Multiple-Choice.