MLDG-Machine Learning Dataset Generator (Pilot Findings) - For Final Year Open-Book Machine Learning Summative Assessment

Aug 2, 2026

Upload by Violet

Following the COVID-19 pandemic, institutions increasingly consider open-book assessments for both proctored and remote settings, but educators face academic integrity, fairness, and equality challenges. The paper presents the development of MLDG, a machine-learning dataset generator that creates student-centred, unique yet fair datasets for a final-year Machine Learning summative course delivered in class and remotely. A two-year pilot study evaluates student views, integrity outcomes, and instructor workload, with encouraging evidence and potential transfer to other computing subjects.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages6

Reading Time15 min

AccessFree to download

Abstract

Introduction

MLDG Development

Pilot Study and Findings

Implications and Future Applications

Technological University Dublin ARROW@TU Dublin | Conference papers | Directorate of Academic Affairs | | --- | --- | | 2023-12-14
MLDG-Machine Learning Dataset Generator (Pilot Findings): For Final Year Open-Book Machine Learning Summative Assessment
Keith Quille
Technological University Dublin, [keith.quille@tudublin.ie](keith.quille@tudublin.ie)
Keith Nolan
TU Dublin, [keith.nolan@tudublin.ie](keith.nolan@tudublin.ie)
Lidia Vidal-Meliá
, UniversitatJaumeI Castelló,Spain, [lvidal@uji.es](lvidal@uji.es)[ ](lvidal@uji.es)See next page for additional authors
Follow this and additional works at: [https://arrow.tudublin.ie/diraacon](https://arrow.tudublin.ie/diraacon)
Part of the Education Commons | | Recommended Citation Quille, K., Nolan, K., Vidal-Meliá, L., & Becker, B. A. (2023) . MLDG-Machine Learning Dataset Generator (Pilot Findings) -For Final Year Open-Book Machine Learning Summative Assessment. HCAI Education and Practice Conference. DOI: 10.21427/AZEE-BZ43 This Conference Paper is brought to you for free and open access by the Directorate of Academic Affairs at ARROW@TU Dublin. It has been accepted for inclusion in Conference papers by an authorized administrator of ARROW@TU Dublin. For more information, please contact [arrow.admin@tudublin.ie](arrow.admin@tudublin.ie), [aisling.coyne@tudublin.ie](aisling.coyne@tudublin.ie), [gerard.connolly@tudublin.ie](gerard.connolly@tudublin.ie), [vera.ki](vera.ki)[lshaw@tudublin.ie](lshaw@tudublin.ie). This work is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 4.0 International License. Authors Keith Quille, Keith Nolan, Lidia Vidal-Meliá, and Brett A. Becker This conference paper is available at ARROW@TU Dublin: [https://arrow.tudublin.ie/diraacon/17](https://arrow.tudublin.ie/diraacon/17) MLDG-Machine Learning Dataset Generator (Pilot Findings) For Final Year Open-Book Machine Learning Summative Assessment Keith Quille School of Enterprise Computing and Digital Transformation, TU Dublin Tallaght, Dublin, Ireland [Keith.Quille@TUDublin.ie](Keith.Quille@TUDublin.ie) Keith Nolan School of Enterprise Computing and Digital Transformation, TU Dublin Tallaght, Dublin, Ireland [Keith.Nolan@TUDublin.ie](Keith.Nolan@TUDublin.ie) Lidia Vidal-Meliá Dept. of Business Administration and Marketing, Universitat Jaume I Castelló, Spain [lvidal@uji.es](lvidal@uji.es) ABSTRACT Following the COVID-19 pandemic, many institutions are considering the continuation of open-book assessments. Students cite many reasons for preferences for open-book assessments, both held in proctored environments and assessments that can be held remotely. Despite the numerous benefits, open-book assessments impose a number of challenges for educators, which is compoundedif the assessment has the option to be taken in class (proctored) and remotely simultaneously. The challenges not only exist around academic integrity but also exist from a student-centred viewpoint of fairness and the equality of the assessment. This paper first presents the development of a tool for generating student-centred datasets (using the MLDG-Machine Learning Dataset Generator) which are unique but fair to allow for both in-class and remote summative assessment for a final-year Machine Learning course, simultaneously. Second, a study was conducted over a two-year period to trail the tool and assessment environment, where it reported positive findings from a student’s viewpoint, academic integrity viewpoint and the workload from course instructors (as the tool also generates an information file to reduce grading time). The results are encouraging and provide evidence that MLDG and the assessment approach used to promote a fair assessment with academic integrity for Machine Learning students. This tooland assessment approach could also be applied in other subjects such as databases thus having value outside of Machine Learning courses for the Com

FAQs

What outcomes were evaluated in the two-year pilot study?

The study reports findings from students, academic integrity perspectives, and instructor workload, including reduced grading time due to an accompanying information file.

Aug 2, 2026

Upload by Violet