Learner profiles inthe Higher Educational Context
C. Skourlas1, C. Sgouropoulou1, P. Belsis1, G. Pantziou1,
C. Sfikas1, N. Fosses1
1Department of Informatics, Technological Educational Institute of Athens, Greece
Tel: +302105910974, Fax: +302105910975
E-mails: {cskourlas, csgouro, pantziou}@teiath.gr, ,
National and institutional policies in Greece intensively foment the reformation of Higher Education (HE) in order to build upon the potential of new Information and Communication Technologies (ICT). The integration of e-learning approaches currently constitutes a high priority issue for tertiary educational institutions. It is important however to take into account the different learner profiles encountered in the higher educational context. Our work emphasizes on issues concerning the design of accessible, adaptive,usable, web-based lessonsfor learners with the cognitive disability of dyslexia.
1.Introduction
According to the ACM/IEEE curricula [1], ‘students need to be able to develop conceptual and physical models, determine methods appropriate for providing efficient solutions to a given problem, and be able to select and implement appropriate solutions that reflect suitable constraints, including scalability and usability’.The need for standardizationof expected learning outcomes and the adoption of such principleshas become of strategic importance for the evolution of the Greek Higher Educational System [2].Addressing this need assumes the treatmentof several hindering factors, one of the most critical being the low ratio of the permanent teaching staff in respect to the continuously growing number of students [3], especially in the case of Technological Higher Education.Hence, the driving forces behind our efforts have been the need for standardization and co-operation, especially due to the lack of human teaching resources, and the exploitation ofthe ‘huge’ potential of new Information and Communication Technologies (ICT) [2] [4].
The ICT impact has undoubtedly fomented the blended learning paradigm and the introduction of online learning as a vital component to the emerging higher educational setting. The integration of multimedia features in the teaching content augmented with the application of multimedia enhanced teaching methodologies [4], [5]arises as an interesting challenge in modern Academic Education schemes. Solid body of theory and practice is already established in the context of distance tutoring and designers and tutors have to choose among and apply various instructional strategies and material. In particular, Belsis et al. [4]havestressed the need for multimedia enhanced Content Deployment and the implied benefits, which are related to the introduction of Computer Based Training techniques, as a part of the traditional course.
Obviously, problems related to the arrangement of distance learning experimentsexist,they could, however, be tackled as a strategy that encompasses specific processes, methods, and tools related to the Software Engineering paradigm and the Business Process Modeling approach [6]. Such a strategy highlightsthe Learner requirementscapturing and analysis activity as a critical task for the overall success of the endeavor. This approach puts the Learner at the centre, in order to improve the understanding of the learning target audience as a whole, but alsoto enable the capturing and examinationof the particular needs of each one of the different learning groups expected totake part into a learning experience.A number of subsequent activities must be co-operatively conducted for developing the instructional multimedia material and incorporating the appropriate educational technology [4] (defining goals, objectives and components of the multimedia material, instructional material to be covered, constraints and assessment procedures).
Combining efficient learner requirements elicitation with personalization techniques throughout the design and delivery of the instructional material is an extremely important aspect forprovidingadaptation to the needs and interests of individual learning groups.
Learners with the cognitive disability of dyslexia constitute a learning group which can be greatly benefited by such implementations. Dyslexia is a specific learning difficulty that actually affects reading and writing / spelling and is caused by problems in the phonological system of language processing.
This paper discusses aspects of our work concerning requirements analysis and personalization techniques for the design of adaptive, web-based, multimedia learning material for dyslexic learners in a higher educational context. More specifically, section 2 reviews the state-of-the-art technology and systems for (web) personalization. Section 3 briefly describes the design and implementation of adaptive learning environments for dyslexic learners. Section 4 presents the results of an initial requirement analysis, while in the last section conclusions and future directions are discussed.
2.Personalization and user model
Personalization is simply defined as the process of making information systems adaptive to the needs and interests of individual users. Using the concept of the Web-based information systems we can define Web personalization in the same way. Typically, web personalization concerns data collection about the users, analysis of these data, and retrieval of the suitable data for the specific user at the suitable time [7].
Personalization can be achieved with the use of a separate personalization server of multimedia educational material that makes use of various types of adaptive personalization: (a) personal user statistics, (b) stereotype modeling, and (c)community modeling. Each of the types requires the acquisition and maintenance of a different user model, which is achieved with the use of statistical analysis and machine learning methods [10].
Three steps are important for successful personalization [11, 12]:
1.Collection of useful information about the users and their interests.
2. The collected data are processed to discover interesting patterns and, create user models. Individual learners are clustered and modeled according to their interests and abilities.
3. New educational material to be presented in the learneris chosen, together with the order of presentation using filtering (and ranking) techniques.
Instead of using the expensive Content-based filtering which is based ondata preprocessing and analysis, the personalization server can use collaborative filtering to group the users into communities according to common characteristics and interests.
Individual user (learner) model may contain personal information about the users, as provided during the registration and information related to the description of sources and categories.Weight parameterscan be defined based on the frequency at which the user chooses the particular source or category for new educational material [10].
Stereotypes are similar to personal user models, but they accumulate frequency statistics for all users with the same personal characteristics [10].
User communities are also aggregate models, but they are not predefined and do not contain personal information about the users. They are constructed with the use of machine learning algorithms. Example of such a machine learning algorithm is Cluster Mining [10, 12], which discovers patterns of common behavior by looking for all fully connected sub-graphs (cliques) of a graph that represents the user's characteristic attributes.
3.Dyslexic learners
In our work as educators, we see people with disabilities and learning difficulties as learners and we put substantial effort in understanding and capturing the individual learner’s requirements. Within this project, we have been concerned with reviewing the state of the art on accessibility issues and have studiedhow appropriate software tools could be used within the mainstream learning environment in order to enable learners to develop their skills and address their needs at their own pace. Finally, we have focused on identifying the ‘learner (dyslexic user) requirements’.
3.1 Design guidelines for dyslexic people
Considerable work has been undertaken in the context of the World Wide Web Consortium's Web Accessibility Initiative (WAI) in order to make the use of the web easier for people with disabilities. WAI accessibility documents include guidelines regarding development andaccessibility features of web sites, browsers and authoring tools. Emphasis is given to the translation of the documents into different languages, the visual appearance of pages to the needs of print-disabled readers and the possibility of speech synthesis for the text being read.
The most important documents of the Web Accessibility Initiative (WAI) are the following:
•The Web Content Accessibility Guidelines (WCAG).
•The Authoring Tool Accessibility Guidelines (ATAG).
•The User Agent Accessibility Guidelines (UAAG).
•The Evaluation and Report Language (EARL).
•The Accessibility Information for Specific Technologies (XML, SVG, SMIL, and other specific technologies).
The British Dyslexia Association for dyslexic readers also proposes selected tips for fonts, colours, fonts’ size, background, presentation style (e.g. characters per line, line spacing, margins, use of bold / italics, use of bullets), etc.
The Centre for Educational Technology Interoperability Standards (CETIS) has also presented some principles and tips aiming at enhancing readability, accessibility, and customization of web pages for people with dyslexia.
3.2Assistive technology
Assistive technology can be used to provide the meansto support individual users to work around reading, writing / spelling, and learning difficulties. Software in common use for supporting dyslexic people can be classifiedinto the following categories:
1)Literacy Teaching integrated environments (e.g. Reading Rockets of the RocketReader Pty Ltd, Kurzweil 3000 of the Kurzeil Educational Systems)
2)Text to Speech (TTS) software (e.g. Ekfonitis+ of the ISLP, Loquendo TTS of the Loquendo Co)
Towards the integration of individual tools, the AGENT-DYSLconsortiumhas proposed a system which can combine pedagogical and technological efforts ‘to monitor the engagement of the (dyslexic) learner and to aid in the identification of individual learning needs’ [8]. The system aims to be both enabling (using text-to-speech conversion programs and spell-checkers) and instructional (using controlled, structured presentation of reading) and supports inclusive learning [9] in Accommodative Learning Environments. According to the objective of AGENT-DYSL program the development of Accommodative Learning Environments includes: generating appropriate user profiles, adapting provision of services and presentations according to the profile of user, and using semantics and knowledge to monitor context of learning.
4.Presentation and discussion of the results of the User (Learner) Requirement analysis
Our effort towards an adaptive, web-based system capable of providing personalized multimedia learning material in order to address particular learner subsets (stereotypes) and short-term individual preferences has been conditioned by the following principle [8]: ‘On the basis of the user profile and performance record, it is desirable that the system provides a range of features that will support its use within accommodative learning environments. This information would enable personalisation of the presentation of learning materials and course texts’.
In accordance with this principle and based on the experience of experts in the field, design guidelines for dyslexic people, and review of existing software tools, the following user requirements have been captured and are briefly presented in the rest of the section.
4.1Type of the educational material (documents) and User Interface
1)The system must support access, use, and handling of educational material. Supported formats for text can include, at least, doc, and PDF documents. Multimedia content could be enhanced by multimedia features integration e.g. sound, figures, diagrams, and videos.
2)The layout of the displayed document should be simple and adjusted.
3) The document’s content should appear in an appropriate format using large letters, appropriate fonts, etc.
4) Downloading and storage of documents should be possible.
5)User friendly, simple, dyslexia sensitive GUI. Multilingual GUI support should be also provided.
4.2Personalization and adaptive features
Significant information about learners can include:
a) Age, first Enrollment date / semester, class.
b)For specific courses there are Prerequisite modules. Therefore, examination marks for these modules, can give additional information about the learner’s profile.
c) Training, professional experience.
Such information is useful for classifying the students into groups (e.g. first class students, working students, students with special needs).
Apart from the understanding of the target audience (learner), the designers / tutors must answer questions as the following ones:
What the learner is expected to learn? What s/he should be able to accomplish? How long it should take?
4.2.1 Text presentation
1) The choice of style is part of the system’s ‘personalization’. Text reformatting must be supported. Dyslexic user can change the font type, font size, font colour, line spacing, etc.
2) Highlighting. The system should support the style of highlighting employed by well known products (e.g. ReadOn).
4.2.2. User profiles
The collection of data about the users and their interests is performed explicitly, through form-filling, and implicitly, through the logging of usage data.Machine learning methods can be used to create adaptive user models that capture changes in the user’s interests [10 - 12].
For each user, the system should maintain a ‘user profile’ which contains information about the performance of the user and user preferences.
The following methods can be used for the assignment of the initial user profile [10]:
•A profile can be manually built based on prior knowledge about the student.
•A profile can be automatically built based on an 'assessment'.
The profile will be updated when the learner uses the system.
As part of the personalization of the system to each individual user, several features can be tuned based on the profile of the user. The following features can be the subject of adaptation [8]:
•Set-up of the system
•Font type, size and colour.
•Speed of highlighting
•Text analysis
5.Conclusions and future activities
In this paper we have focused on students of Higher Education. We have been concerned a) with the user (learners) requirements and b) with the design of educational course material. Our work emphasizes on students with disabilities and learning difficulties.
We have also studied how software products can enable the learners to cover their needs. The extraction of design guidelines for the presentation of the educational course material was based on the principles, guidelines, and tips of international organizations, and initiatives for people with disabilities.
We proposed the use of a personalized server of multimedia, educational course material and discussed the importance of supporting personalization techniques based on user models and stereotypes. Of primary interest is the experimentation with user communities in order to try to cover needs of learners.
Future research should focus on the evaluation (effectiveness) of the proposed approach. We intend to examine other user (learner) categories; we also intend to establish a framework not only for dyslexic people but also for people with other disabilities (vision or hearing disabilities). In addition we intend to perform an analysis aimed at facilitating other groups with specific characteristics and needs (e.g. working students).
Trying to specify and apply new innovative features we shall focus on:
1) Automatic adaptation of the document presentation when a change in the user profile takes place.
2) Analysis of text, taking into account the profile of the student. If parts of the text are “difficult” then the presentation will be adapted accordingly.
3) Suggestions(based on the user’s profile and learners’ ‘performance’) for ‘further reading’ and / or exercises that will help the student.
Acknowledgements
This work is co-funded by 75% from E.U. and 25% from the Greek Government under the framework of the Education and Initial Vocational Training II, programme “Supporting the Informatics Studies in the T.E.I. of Athens”. We would like to thank our advisors in the fields of educational technology and dyslexia Associate ProfessorAntonis Symvonis and Helen Mitropouloufor their valuable contribution to determine the needs for a system that will support learning in an accommodative environment.
References
[1]ACM/IEEE: Computing Curricula 2001, vol. II, Computer Science, 2001.
[2]Alevizos T., Skourlas C., “Database curricula in Higher Informatics Education”, Proceedings of Workshops on Computer Science Education, Jan. 2004, pp. 11-15, Nis, Serbia.
[3]Kaburlasos, V., Marinagi, C., Tsoukalas V. “PARES: A Software Tool for Computer-Based Testing and Evaluation Used in the Greek Higher Education System”, 4th IEEE International Conference on Advance Learning Technologies (ICALT 2004) 30 August – 1 September 2004 Joensuu, Finland
[4]Belsis P., Pantziou G., Skourlas C., Varnas J. “Supporting Academic Courses through Multimedia Enhanced Content Deployment, International Conference on Multimedia, Image Processing and Computer Vision, Madrid March 2005
[5]Rossiou E., Pantziou G., “The changing model of teaching and learning multimedia”, Proceedings of Workshops on Computer Science Education, Jan. 2004, pp. 31-34, Nis, Serbia
[6]Alevizos T., Belsis P., Pantziou G., Skourlas C., Tsoukalas V., Varnas J., Supporting Computing Curricula through Distance Learning and Multimedia enhanced content deployment - the case of Greek Technological Educational Institutes, Wseas 2006
[7]Pierrakos D., Paliouras G., Papatheodorou C., and C.D. Spyropoulos, “Web Usage Mining as a Tool for Personalization: A Survey, User Modeling and User-Adapted Interaction”, v. 13, n. 4, pp. 311-372, 2003.
[8]Agent-DYSL project, Deliverable 1.1. State of the Art and Requirement Analysis Report, 28-12-2006
[9]Tomlinson, J. (1996) Inclusive Learning: Principles and Recommendations – a summary of the findings of the Learning Difficulties and Disabilities Committee, Coventry: Further Education Funding Council.
[10] Paliouras G., Mouzakidis A., Ntoutsis C., Alexopoulos,A., Skourlas C., “PNS: Personalized Multi-Source News Delivery”, KES 2006 conference
[11] Kobsa A., Koenemann J.,Pohl W., “Personalized Hypermedia Presentation Techniques for Improving Online Customer Relationships”. The Knowledge Engineering Review 16 (2), pp. 111-155, 2001.
[12] Paliouras G., Papatheodorou C., Karkaletsis V., and C.D. Spyropoulos, 2000, Clustering the Users of Large Web Sites into Communities, Proceedings of the International Conference on Machine Learning (ICML), pp. 719-726, Stanford, California.