Genetic algorithm with temporal logic for automated markers

dc.contributor.advisorBosman, Anna S.
dc.contributor.coadvisorMarshall, Linda
dc.contributor.coadvisorNils, Timm
dc.contributor.emailu19089296@tuks.co.za
dc.contributor.postgraduateRedelinghuys, Francois Jacobus
dc.date.accessioned2025-07-14T10:59:43Z
dc.date.available2025-07-14T10:59:43Z
dc.date.created2025-09
dc.date.issued2025-07
dc.descriptionDissertation (MSc (Computer Science))--University of Pretoria, 2025.
dc.description.abstractAutomated assessment is essential for large-scale programming courses, as it addresses the challenges of evaluating correctness and providing feedback efficiently. Traditional methods of automated assessment rely on hand-crafted test cases, which are time-intensive and lack scalability. Automated test generation methods, such as random input generation or genetic algorithms combined with code coverage metrics, offer alternatives, but often fail to capture the complexity required for educational assessments. To address this, a novel approach that integrates genetic algorithms with linear temporal logic (LTL) formulae is proposed. LTL properties, commonly used in model checking, formalise correctness criteria for programming assignments. The proposed method evolves test cases designed to maximise violations of the LTL properties in student submissions, determining correctness and generating personalised feedback by providing examples of failing test cases. Experiments \replaced{were}{are} conducted to evaluate the approach in terms of sensitivity to hyperparameter settings, and effectiveness relative to the established techniques, such as random input generation and code coverage-based genetic algorithms. Results demonstrate that combining genetic algorithms with LTL properties enhances automated assessment accuracy and feedback quality, offering a solution to the problems faced by automated assessment in large educational settings for both students and lecturers.
dc.description.availabilityUnrestricted
dc.description.degreeMSc (Computer Science)
dc.description.departmentComputer Science
dc.description.facultyFaculty of Engineering, Built Environment and Information Technology
dc.description.sdgSDG-04: Quality Education
dc.identifier.citation*
dc.identifier.doihttps://doi.org/10.25403/UPresearchdata.29554802
dc.identifier.otherS2025
dc.identifier.urihttp://hdl.handle.net/2263/103338
dc.language.isoen
dc.publisherUniversity of Pretoria
dc.rights© 2024 University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria.
dc.subjectUCTD
dc.subjectSustainable Development Goals (SDGs)
dc.subjectGenetic algorithm
dc.subjectTemporal logic
dc.subjectAutomated marking
dc.titleGenetic algorithm with temporal logic for automated markers
dc.typeDissertation

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Redelinghuys_Genetic_2025.pdf
Size:
63.66 MB
Format:
Adobe Portable Document Format
Description:
Dissertation

License bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description: