Naslov
Utilizing Advanced Software Solutions for Academic Work Integrity and Learning Validation

Autori
Mutka, Alan ; Nižetić, Josip ; Žagar, Martin ; Mihaljević, Branko

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Proceedings of EDULEARN21 Conference / - : International Academy of Technology, Education and Development (IATED), 2021, 6832-6837

ISBN
978-84-09-31267-2

Skup
13th International Conference on Education and New Learning Technologies (EDULEARN21)

Mjesto i datum
Online, 05.07.2021. - 06.07.2021

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Work integrity, learning validation, virtual proctoring, plagiarism, synchronous assessment

Sažetak
The COVID-19 pandemic has caused an educational crisis worldwide, starting from unequal access to digital technology and the internet, suitable home learning environments, and family support to educators and students' preparedness and ability to work on an online platform. The fast transition to online remote delivery of teaching and learning has also seriously influenced academic integrity concerning student work, including, but not limited to, homework assignments, exams, online labs, etc. The latest software solutions for virtual proctoring provides support with remote supervision of an exam and often uses AI to track the head-body pose, gaze orientation, and computer activities such as opened windows and tabs. Most of these solutions apply restrictions - for example, full-screen mode and limited to no access to other browser windows - as the primary protection tool ; which is often not adequate if students are required to use additional software during an exam and access online documentation. Finally, the provided solutions are mainly focused on a student's external behavioural activities during an exam. At present, there are no solutions for tracking the academic work integrity of student assignments and exams, which are not created and supervised in real-time. This research paper presents a novel approach that uses AI solutions to track and examine task creation in real-time. Currently, we are focused on programmed source code and simple text creation ; however, this solution can be extended to any digitally written assessments. The fundamental concept is to track and record the student's input events (keyboard, mouse, clipboard) recorded in real-time during in-class exercises. The catalogued data are subsequently analyzed, and a visual representation of the synchronously created work is presented and validated. Furthermore, a students’ progress and/or improvement can be obtained, evaluated, and validated. The proposed software solution has been successfully tested and validated on the Web and Mobile Computing and International Business programs at Rochester Institute of Technology Croatia.

Izvorni jezik
Engleski

Znanstvena područja
Računarstvo

POVEZANOST RADA