Use this URL to cite or link to this record in EThOS:
Title: Presence verification for summative e-assessments
Author: Apampa, Kikelomo Maria
ISNI:       0000 0004 2703 9327
Awarding Body: University of Southampton
Current Institution: University of Southampton
Date of Award: 2010
Availability of Full Text:
Access from EThOS:
Access from Institution:
Influenced by information technology advances, the assessment process has begun to make its way out of the traditional classroom into online environments. The online summative assessment is a high-stake examination which counts towards a final course mark. Thus, as a result of the important consequences of such summative tests, security measures are put in place to ensure that only the ‘right’ students are assessed. However, the identity-authentication model adopted for user security is susceptible to impersonation challenges. This thesis introduces the concept of presence verification as an essential extension to the existing identity-authentication user security model. The presence security goal is aimed at ensuring that the correctly authenticated student at the start of a test is the same student throughout the test session. Thus, verifying a student’s presence beyond the initial login procedure minimises the impersonation threats. In order, to embrace the gains of ensuring presence during summative e-assessments, a blob-analysis solution which follows an object tracking approach is proposed. The design of the blobbased presence verification (BlobPV) system involves video processing techniques which can be used to detect, verify and classify a student’s presence status in the test environment. Thereby, indicating the likelihood of acceptable or unacceptable activities. Experiments were carried to demonstrate the feasibility of a blob-based presence verification system in summative test environments. Additionally, the BlobPV system was evaluated to determine the accuracy of correctly classifying a student’s presence status. For each experiment and evaluation, the methods and results are described. The results clearly state that, such an approach would significantly improve the detection rate of impersonation attempts during online summative assessments.
Supervisor: Wills, Gary ; Argles, David Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: QA75 Electronic computers. Computer science