Research Article
Question Design Using NLP
@INPROCEEDINGS{10.1007/978-3-031-81573-7_8, author={Maty Sene Kane and Alassane Diop and Zinflou Arnaud and El Hadji Mamadou Nguer}, title={Question Design Using NLP}, proceedings={Towards new e-Infrastructure and e-Services for Developing Countries. 15th International Conference, AFRICOMM 2023, Bobo-Dioulasso, Burkina Faso, November 23--25, 2023, Proceedings, Part II}, proceedings_a={AFRICOMM PART 2}, year={2025}, month={2}, keywords={NLP generate questions automatically examination management system}, doi={10.1007/978-3-031-81573-7_8} }- Maty Sene Kane
Alassane Diop
Zinflou Arnaud
El Hadji Mamadou Nguer
Year: 2025
Question Design Using NLP
AFRICOMM PART 2
Springer
DOI: 10.1007/978-3-031-81573-7_8
Abstract
Cheikh Hamidou KANE Digital University (ex UVS) has more than 60,000 students enrolled in 46 tracks containing more than 1,000 courses, the assessment of which requires the preparation of nearly 2,000 exam topics per year. Since its opening in 2013, it is estimated that nearly 11,000 topics have been prepared, and this number is growing (see [4]). In addition to these numbers, on the one hand we have the lack of an examination management system (EMS). On the other hand, we see enormous difficulties in receiving assessment topics at the teacher level, given the large number of topics a teacher has to design. We propose a framework called JEMS (Jolof Examination Management System) that allows for declarative expression and evaluation of topics close to their design. In order to be able to generate questions automatically, we have made use of other models such as natural language processing based models. This framework aims to automate evaluation questions from an existing question bank for a(\textrm{CE})(Constituent Element). Automating evaluations in this context poses a number of scientific challenges that constitute the contributions of the presented work:
The implementation of a file template in csv format.
Question models that identify entities in CEs and generate statements against the natural language processing model
