FACILITATING THE AUTOMATED ASSESSMENT OF ENGINEERING COURSES
1 Umited Arab Emirates University (UNITED ARAB EMIRATES)
2 Cortex Business Solutions (CANADA)
About this paper:
Conference name: 12th annual International Conference of Education, Research and Innovation
Dates: 11-13 November, 2019
Location: Seville, Spain
Abstract:
Introduction:
A massive open online course (MOOC) is expected to be taken by a large number of students possessing a variety of strengths and weaknesses, learning styles, and cultural and educational backgrounds [1]–[3]. The hundreds of current MOOCs face the challenge of not being recognized as ‘regular’ courses mainly because the course assessment is done in an unsupervised, cheating-prone environment. To alleviate this problem, a few solutions include: constricting the test-times, not repeating questions, using a large question database, etc. [4] The automated assessment of many electrical/computer engineering (ECE) courses can be enhanced significantly by using large databases comprising unique questions-and-answers – doing so is the aim of this work.
Automated question-and-answer generation:
At present, a variety of university-level ECE courses are offered as MOOCs [5], [6]. We propose an automated tool (named QAgen) for facilitating the automated assessment of these courses. QAgen is based on open-source Octave language and is able to create 1000’s of questions and answers for different courses, e.g., Digital Logic, Hardware Description Language Based Design, Computer Architecture, Circuit Testing, etc.
The question creation for many ECE courses entails one or more of these tasks: the automatic generation of tables of varying sizes, the creation of properly formatted equations with multiple variables, and the drawing of arbitrary logic diagrams. Among these tasks, the automated drawing of diagrams is significantly harder than the creation of the tables or the equations.
QAgen generates all of the question-related ‘outputs,’ which include not only the question text and the related diagrams but also the answer text. QAgen is fully configurable by using different parameters, such as circuit element types, circuit size, etc.
Results and Conclusion:
As a proof-of-concept, a large number of questions were generated for three different traditional/classroom-based courses during the Spring 2019 semester (CENG205, CENG221, and CENG513). A few examples of the questions uploaded on BlackBoard (a well-known LMS) are shown in the paper. The examples include questions about logic equations, logic diagrams, Karnaugh maps, and truth tables for our Digital Design & Computer Organization course. At this stage, the questions were provided to the students only for practice, i.e., their scores were not included in their actual course grades. The students were asked to fill in a survey about the questions. The majority of students used the questions for practice before the actual (graded) test/examination. Most of the students also felt that the questions helped them prepare for the tests/examinations. The availability of the answers to the questions were overwhelmingly favored by the students. A large number of students were more confident in problem solving after going through the questions. Finally, the students feel that their grades improve when they practice with the question-answer sets. Our initial assessment of the QAgen tool has provided us with very encouraging results and hence the potential of extending the tool’s usage to MOOC(s). The principles used herein are expected to benefit other courses as well, provided they require the creation of diagrams with connected geometric objects, code-based problems, equations and their solutions, etc.
References:
[1] W. Hao et al., “Research on Blended Teaching Reform and Innovation Strategy Based on MOOC Education,” in 2018 13th Int. Conf. Comput. Sci. Educ., 2018, pp. 1–4.
[2] M. Lu, et al., “A Review of the Recent Studies on MOOCs,” in 2018 13th Int. Conf. Comput. Sci. Educ., 2018, pp. 1–5.
[3] H. S. H. Ip, et al., “Design and Evaluate Immersive Learning Experience for Massive Open Online Courses (MOOCs),” IEEE Trans. Learn. Technol., pp. 1–10, 2018.
[4] Z. Posner, “What is Adaptive Learning Anyway?” 2019. https://www.mheducation.com/ideas/what-is-adaptive-learning.html
[5] “Coursera” 2019. https://www.coursera.org/
[6] “Udacity” 2019. https://www.udacity.comKeywords:
Massive open online course (MOOC), engineering education, distance learning, e-learning, automated assessment, test-bank, logic circuit, logic diagram