DIGITAL LIBRARY
ANALYSIS OF ALGORITHMS FOR GENERATING TEST QUESTIONS IN E-TESTING SYSTEMS
University of Library Studies and Information Technologies (BULGARIA)
About this paper:
Appears in: EDULEARN19 Proceedings
Publication year: 2019
Pages: 1714-1719
ISBN: 978-84-09-12031-4
ISSN: 2340-1117
doi: 10.21125/edulearn.2019.0498
Conference name: 11th International Conference on Education and New Learning Technologies
Dates: 1-3 July, 2019
Location: Palma, Spain
Abstract:
The introduction and use of the electronic test not only as a means of verifying knowledge but also as a means of self-learning has led to the need to generate multiple training tests. Besides the well-known learning management systems (LMS), a lot of applications have been created in the space that offers the generation of training tests. Each one offers a set of algorithms to generate different types of test questions. In addition, each system offers a specific random selection mechanism for both the answers to a question and the order of the test questions in the final test. We make analysis and after that, it was found that different systems (LMS or specialized test systems) have, besides similar and specific ones, the algorithms themselves, but in each one of this system, the question is created/generated one by one. This takes a significant amount of time for teachers to create their own questions and then upload them into their chosen system. This article analyzes which of the existing systems offers the best opportunity to prepare test exams individually for each learner. For this purpose, the capabilities of well-known LMSs and less used test system ones have been examined and analysed. This analysis will be the basis for preparing a model for the accelerated generation of test questions and grouping them into individual tests that save time for teachers.
Keywords:
Test, algorithms, LMS, analyses.