Abstract
<p>Student testing and knowledge assessment is a significant aspect of the learning process. In a number of cases, it is expedient not to present the exact same test to all learners all the time (Pritchett, 1999). This may be desired so that cheating in the exam is made harder to carry out or so that the learners can take several practice tests on the same subject as part of the course.</p><p><br />This study presents an e-testing platform, namely PARES, which aims to provide assessment services to academic staff by facilitating the creation and management of question banks and powering the delivery of nondeterministically generated test suites. PARES uses a conflict detection algorithm based on the vector space model to compute the similarity between questions and exclude questions which are deemed to have an unacceptably large similarity from appearing in the same test suite. The conflict detection algorithm and a statistical evaluation of its accuracy are presented. Evaluation results show that PARES succeeds in detecting question types at about 90% and its efficiency can be further increased through continuing education and enrichment of the system’s correlation vocabulary.<br /><br /></p><p> </p>
Highlights
In recent years, e-learning has made significant progress in every way that can be measured
The latest improvement in PARES, which is the main subject of this paper, concerns the integration of information retrieval (IR) techniques to identify conflicting questions in the question banks and prevent their mutual inclusion in the same test instance
To evaluate the PARES efficiency of finding conflicting questions, 103 exam questions were submitted for three higher education courses: 45 on Telematics, 32 on Distance Education, and 26 on Teaching Information Technology
Summary
E-learning has made significant progress in every way that can be measured. The latest improvement in PARES, which is the main subject of this paper, concerns the integration of information retrieval (IR) techniques to identify conflicting questions in the question banks and prevent their mutual inclusion in the same test instance This functionality is a specialized case of the search problem and uses keywords for each question to compute the similarities between questions using the cosine function in the vector space model (Salton et al, 1975). To refer to the multitude of different types of information articles, we will be using the general term document While initially it seems that finding the required information is the only task performed by an IR system, today’s large information corpora present another, not significantly easier, challenge: how to ascertain which of the multitude of search results better corresponds to the input data.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: The International Review of Research in Open and Distributed Learning
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
