Title: An effective evaluation system to grade programming assignments automatically

Authors: Md. Afzalur Rahaman; Abu Sayed Md. Latiful Hoque

Addresses: Department of Computer Science and Engineering, Hamdard University Bangladesh, Gazaria, Munshiganj, Bangladesh ' Department of Computer Science and Engineering, Bangladesh University of Engineering and Technology, Dhaka, Bangladesh

Abstract: For the last decades, programming courses are being taught in nearly every educational sector. Students are now more likely to use an e-learning platform compared to traditional system because of lower internet costs, remote access, and faster communication facilities. For a programming course studied in both manual and e-learning platforms, instructors evaluate huge number of programming assignments which is cumbersome since programming assignment checking requires higher accuracy. To mitigate the instructor's job with a satisfactory level of confidence, we developed a system using TF-IDF, generalised graph isomorphism, and token edit distance. The testing outcome of the system reveals a higher level of congruence with a human expert grader.

Keywords: generalised graph isomorphism; GGI; partial grading; solution feature; solution matrix; TF-IDF; token edit distance.

DOI: 10.1504/IJLT.2022.127194

International Journal of Learning Technology, 2022 Vol.17 No.3, pp.267 - 290

Received: 30 Sep 2021
Accepted: 25 Jan 2022
Published online: 25 Nov 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article