Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.572201
Title: Personality representation : predicting behaviour for personalised learning support
Author: Wood, Dawn Helaine
Awarding Body: University of Hull
Current Institution: University of Hull
Date of Award: 2010
Availability of Full Text:
Access through EThOS:
Access through Institution:
Abstract:
The need for personalised support systems comes from the growing number of students that are being supported within institutions with shrinking resources. Over the last decade the use of computers and the Internet within education has become more predominant. This opens up a range of possibilities in regard to spreading that resource further and more effectively. Previous attempts to create automated systems such as intelligent tutoring systems and learning companions have been criticised for being pedagogically ineffective and relying on large knowledge sources which restrict their domain of application. More recent work on adaptive hypermedia has resolved some of these issues but has been criticised for the lack of support scope, focusing on learning paths and alternative content presentation. The student model used within these systems is also of limited scope and often based on learning history or learning styles. This research examines the potential of using a personality theory as the basis for a personalisation mechanism within an educational support system. The automated support system is designed to utilise a personality based profile to predict student behaviour. This prediction is then used to select the most appropriate feedback from a selection of reflective hints for students performing lab based programming activities. The rationale for the use of personality is simply that this is the concept psychologists use for identifying individual differences and similarities which are expressed in everyday behaviour. Therefore the research has investigated how these characteristics can be modelled in order to provide a fundamental understanding of the student user and thus be able to provide tailored support. As personality is used to describe individuals across many situations and behaviours, the use of such at the core of a personalisation mechanism may overcome the issues of scope experienced by previous methods. This research poses the following question: can a representation of personality be used to predict behaviour within a software system, in such a way, as to be able to personalise support? Putting forward the central claim that it is feasible to capture and represent personality within a software system for the purpose of personalising services. The research uses a mixed methods approach including a number and combination of quantitative and qualitative methods for both investigation and determining the feasibility of this approach. The main contribution of the thesis has been the development of a set of profiling models from psychological theories, which account for both individual differences and group similarities, as a means of personalising services. These are then applied to the development of a prototype system which utilises a personality based profile. The evidence from the evaluation of the developed prototype system has demonstrated an ability to predict student behaviour with limited success and personalise support. The limitations of the evaluation study and implementation difficulties suggest that the approach taken in this research is not feasible. Further research and exploration is required –particularly in the application to a subject area outside that of programming.
Supervisor: Ishaya, Tanko Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.572201  DOI: Not available
Keywords: Computer science
Share: