Transactions Template JOURNAL OF ENGINEERING RESEARCH AND TECHNOLOGY, VOLUME 3, ISSUE 3, SEPTEMBER 2016 66 Designing Domain Model For Adaptive Web-based Educational System According to Herrmann Whole Brain Model Mohammed Ahmed Ghazal 1 , Nor Azan Mat Zin 2 , Zurina Muda 2 1 University College of Science and Technology, Khan Younis – Gaza Strip, Palestine, m.ghazal@cst-kh.edu.ps 2 University Kebangsaan Malaysia, Selangor, Malaysia Abstract: Educational materials represent a domain model of Adaptive Web-based Educational System (AWBES). However, these materials should be designed to cover the differences of learners‟ preferences. Herrmann Whole Brain Model (HWBM) is a reliable Learning Style (LS) model which can be used to extract the learner‟s preferences in educational environment according to brain structure of learner. In this paper, the learning materials of an essential programming language course (C++) are organized to cover all learners‟ differences according to their brain dominance. The learning materials were described and classified by instructional metadata to fit the preferences of four brain quadrants (rational, organizational, interpersonal and intuitive) within diverse learning objects. The main advantage of this approach is that it is not related to particular type of learners, but it covers the different learners according to their brain-structure. The system which could apply this model can be used to detect the learner preferences dynamically and thus personalize the learning materials within Web-based Educational System (WBES). Index Terms— Domain Model, Adaptive Web-based Educational System, Herrmann Whole Brain Model, Learning Style, Learner Model. I INTRODUCTION Learning Object (LO) represents any digital learning con- tent which can be used to develop the learning environment in order to support learning process. The main importance of advent the learning objects is the need for re-using learning materials which are authored by the teacher or another per- son. Currently, most of the researches in learning systems tend to enhance the machine-driven and automate of gener- ating learning objects. For instance, the lesson is presented to study by the student through gathering a set of learning objects automatically. However, the most challenge is that how the learning objects and courses can be used to person- alize the content presentation to the learner through adequate matching between learner preferences and most related learning objects. But today, achieving accurate adaptivity between the learners and their related contents of learning environment is not really possible. The automatic adaptivity requires further educational metadata to carry a useful in- formation about each learning object[1]. IEEE Learning Object Metadata (LOM) is the most wide- ly accepted and used standard which is made to describe the learning objects from the very practical needs for assem- bling different learning materials from reusable learning objects [2]. This standard identified 76 different attributes to support the interoperability and adaptivity between learner and the domain of learning objects [3]. A metadata field named is used as the most domi- nant attribute that is related to pedagogical and instructional perspectives for educational resources. The possible values of this attribute are: Exercise, Simulation, Questionnaire, Diagram, Figure, Graph, Index, Slide, Table, Narrative Text, Exam, Experiment, Problem Statement, Self-Assessment, or Lecture [4]. Dublin Core Metadata Initiative (DCMI) standard has more broad-range purposes metadata schema which com- prise 15 attributes in the Dublin Core metadata set to de- scribe the wide range of learning objects [5]. DCMI has conducted different activities through working groups, con- ferences, global workshops and educational efforts to identi- fy widespread acceptance of metadata standards. The Dublin Core Metadata Element Set (DCMES) was the first metadata standard which was developed through DCMI as Internet Engineering Task Force (IETF) standard. The DCMES iden- tified different sets of vocabulary to describe the core of information property (e.g., “Title”, “Creator”, “Date” and “Description”) [6]. Furthermore, one of the main challenges in the existing standards is that IEEE LOM failed to represent enough and sufficient level of granularity to describe and identify the instructional part of learning resources [6, 7]. The elements of data related to learning resource type should contain both of technical and instructional information. Therefore, LOM Mohammed Ahmed Ghazal, Nur Azan Mat Zin, Zurina Muda / Designing Domain Model For Adaptive Web -based Educational System According to Herrmann Whole Brain Model (2016) 67 has covered the part of instructional role of learning object (e.g., Exercise, Experiment, Simulation) and the part of technical information for LO which concern their format (e.g., Figure, Graph, Diagram, Table, Slide). However, LOM and other learning object classifications have missed to cov- er several instructional types such as Example, Definition, Terminologies, Theorem, Storytelling, Journaling, FAQ, Drama, and others that are needed for tracking the learners' needs and preferences in the context of holistic learning en- vironment. To overcome this limitation, Gascueña, Fernandez- Caballero [8] proposed a domain ontology to represent and describe the components of learning materials independently through organizing the courses into set of concepts and learning objects to be capable of providing the adaptivity and the reuse of the learning objects. These learning objects were described to cover the diversity preferences of Felder- Silverman Learning Style Model. However, the previous standards and ontologies which are designed according to pedagogical learning theories missed to cover the require- ments of LSs of HWBM [9, 10]. II ADAPTIVE WEB-BASED EDUCATIONAL SYSTEM (AWBES) An Adaptive Web-Based Educational System (AWBES) is a form of online instruction that is used to address the chal- lenges of a WBES [11]. It provides mechanisms to track the learner interactions in order to identify the learner prefer- ences which lead to personalising the design features of a WBES [12, 13]. This system also helps learners accomplish their learning tasks and obtain their required information by adjusting the environment according to their individual dif- ferences and thus, automatically fulfil the learners‟ require- ments [14, 15]. As shown in Figure 1, an AWBES comprises three main components [11]: (1) Learner actions, which track and audit a learner‟s interactions within the design features of WBES in order to derive a learner‟s characteris- tics such as learner‟s preferences and styles; (2) Learner profile, which uses different methods (explicit i.e., question- naire or implicit i.e., prototype) to identify a learner‟s char- acteristics in learner model; and (3) Adaptation methods, which are derived from a learner‟s characteristics in a learn- er model. These learner characteristics are the basic features for developing the adaptation methods of an AWBES [16]. In this research we will focus on learner actions as a main source of identifying learner characteristics implicitly. A Learner Actions Learner actions are used to identify the interaction prefer- ences of learner in the system. The learner behaviour is usu- ally used to describe the real actions of a learner within the system. Therefore, it is considered a more realistic and accu- rate source to build the learner model. There are two ap- proaches of managing behaviour information within the sys- tem. In the first approach, in case of repetition of learner behaviour in the system, the system can translate the con- sistently repetitive behaviour into learning patterns. These patterns can be used to identify the learner‟s real interests and preferences according to his/her real behaviour, and thus, derive more accurate adaptation methods [17]. The second approach is a Cognitive-Science based approach, which focuses on investigating literature in different do- mains, such as the educational and psychological domains. For example, the Learning Style (LS) model was used to gather the prospective relationships between learners and their preferences by analysing the learners‟ interactions within a learning environment. These relationships are rep- resented by predefined learning patterns [18]. Therefore, this research is conducted based on HWBM as a brain-based learning style model [19] Figure 1: The Architecture of AWBES B Herrmann Whole Brain Model Learning Style Learning Style is used to clarify the habitual approach and individual preferences and to organise and represent infor- mation [20]. LS reflects the individual learning preferences that affect how a learner tends to acquire knowledge in the learning process [21]. Keefe in Brown, Brailsford [22] de- fined LS as the “characteristic, cognitive, affective and psy- chological behaviours that serve as relatively stable indica- tors of how learners perceive, interact with, and respond to, the learning environment.”. This research has attempted to apply a brain-based LS model, where, the learner‟s brain structure is the dominant factor in promoting effective learn- ing [23]. Additionally, BECTA and Radwan [24] has showed that the best approach to integrate LS with the most innate and psychological preferences is to exploit LS based on brain-based learning theories. For example, the right hemi- sphere of the brain accommodates creative activities, while Mohammed Ahmed Ghazal, Nur Azan Mat Zin, Zurina Muda / Designing Domain Model For Adaptive Web-based Educational System According to Herrmann Whole Brain Model (2016) 68 the left hemisphere of the brain accommodates logical activ- ities. Furthermore, LS is conceptualised as consistent pat- terns of learning activities that reflect the attitude, prefer- ences, beliefs and motivational orientations of a learner to- wards his/her learning environment [25]. Therefore, incorpo- rating learning patterns of LS models with the design fea- tures of a WBES is useful in linking the identification pro- cess of the LS according to the behaviour of a learner with the system rather than make the identification process static. This research in particular has benefited from using HWBM LSs for modelling the most innate and intrinsic learner pref- erences implicitly and automatically. The HWBM is one of the most reliable and important LS models [26-29]. HWBM is used to extract the most innate and intrinsic learner preferences, which are derived from identifying a learner‟s brain dominance [27]. Furthermore, HWBM is represented by predefined learning patterns. The- se patterns aim to integrate a learner‟s brain dominance with several learning preferences and styles into the features of a learning environment [27, 30]. The HWBM shows that eve- ry learner‟s brain is classified into four brain quadrants [31], where each brain quadrant corresponds to a set of homoge- neous LSs. QA learners can be described as having rational, analytical, logical and theoretical LSs. QB learners can be described as having organising and sequential LSs. QC learners can be described as having interpersonal, emotional, kinaesthetic, expressive and practical LSs. QD individuals can be described as having holistic, intuitive, integrated and synthesising LSs [31]. III DESIGNING CONTENT MODEL FOR ADAPTIVE WEB-BASED EDUCATIONAL SYSTEM (AWBES) Based on the review of the HWBM LS which was conduct- ed in [32], it has been found that learning content is an im- portant part of the WBES design feature particularly when auditing and tracking learner behaviour in a WBES. The content model of a WBES should address the diverse learner requirements according to the HWBM LS. Here, the content model was used to propose an adaptive learner model by identifying learner preferences and LS in the WBES, via analysis of learner behaviour interaction with learning con- tent design features. This section presents a dedicated way of structuring and classifying learning content in a WBES. The learning content should be presented using more descriptive information so that more information can be gained from the behaviour of learners within each aspect of the learning con- tent. A Organising Learning Content of WBES In this study, the online course is the most complex learning object; the smallest learning objects can be represented us- ing different parts of the learning resource including the in- troduction, abstract, image, figure, video, and example. The proposed structure was designed to give learners the main role in a learning process, in the context of traditional educa- tion classrooms or educational systems, as it is based on book taxonomy rather than course taxonomy. Figure 2: Organising of Learning Content in a WBES This research conceptualises learning content using a hierar- chical organisation as shown in Figure 2. Each course con- sists of several modules; each module consists of a set of lessons; each lesson contains a topic or a set of topics; and each topic comprises of several different types of education- al resources represented by fragments. The lowest granulari- ty level comprises the smallest learning objects, which were implemented and stored as a physical file along with its as- sociated metadata. The programming language C++ course was selected for this study. The structure of the learning con- tent was designed according to this hierarchy: (1) the C++ course consists of several modules, where each module co- vers only one subject area; for instance, statements, loops and arrays represent three different modules that demon- strate three different subjects; (2) each module consists of different lessons designed to cover a set of learning objec- tives; for instance, the „for loop‟, „while loop‟ and „do while loop‟ are three different lessons related to the module of loop statements; (3) every lesson has different topics designed to achieve different learning objectives; each lesson comprises different global learning objects such as syllabus, objectives, overview and assessments; (4) every topic aims to achieve one learning objective and comprises different fragments represented by the smallest granularity of learning objects such as introduction, abstract, prerequisites, tests (pre-test or final test), example and other learning objects, which present the concepts in the topic with different styles. According to Popescu [33], this organisation is the most applicable structural way that teachers tend to use when or- ganising their material. Moreover, this organisational man- ner can be used to resolve the following issues: (1) exchang- ing and reusing the learning objects in different manners; (2) Mohammed Ahmed Ghazal, Nur Azan Mat Zin, Zurina Muda / Designing Domain Model For Adaptive Web -based Educational System According to Herrmann Whole Brain Model (2016) 69 tracking the learner‟s interactions with the different types of content learning; and (3) achieving the fine granularity of adaptivity. B Designing Learning Object Metadata Educational metadata was used to add descriptive infor- mation to the learning object. The metadata was applied to facilitate the association between learning objects and learn- er preferences so that learner preferences of learning content could be modelled. For example, Ullrich [7] and Gascueña, Fernandez-Caballero [8] proposed two independent ontolo- gies to represent the educational metadata that associates LS with the most appropriate learning objects. However, the proposed approach by Ullrich [7] is fraught with problems. For example, the ontology that links the metadata with par- ticular dimensions of LS is static and not related to the be- havioural interactions of the learner. It also does not apply implicit techniques in learner modelling. Also, the learning object does not have enough information about the learner. The limitations of the work of Gascueña, Fernandez- Caballero [8], on the other hand, are related to linking learn- ing objects with the Felder-Silverman learning style model. Learning objects are classified into limited categories with- out including significant learning objects, which may be related to other LSs such as communication LOs, help and support LOs, and several fundamental LOs (e.g., definition, objectives, problems, case studies, experiment information, etc.). Therefore, this research has added some extensions to the metadata file to better cover the requirements of the HWBM and the design features of a WBES. These exten- sions aim to enhance previous approaches, including the Dublin Core Metadata Intuitive, Gascueña, et al.‟s [8] in- structional ontology, and Ullrich‟s [7] instructional material. Below are the metadata characteristics of the learning ob- jects used in this research. C General Metadata Characteristics for Learning Object The following metadata characteristics were selected from the standard metadata characteristics that describe learning content: a. title (resource name) → dc:title; b. identifier (refer to resource address e.g., URL) → dc:identifier; c. type (refer genre, nature or form of the content of the resource e.g., service, software, collection, moving, im- age or sound) → dc:type; and d. format (the digital or physical manifestation of the re- source e.g., size, number of pages, and duration) → dc:format. D Educational Learning Object Metadata The hierarchal educational learning objects were used to describe the learning resources, which are related to the learners‟ preferences according to the Herrmann Whole Brain learning theory. The proposed metadata does not de- scribe the learning content. However, it is used to classify the learning content, where each class of metadata refers to a particular instructional role and its related learning resource [34]. The instructional role is a kind of protocol specification that identifies characteristics and behaviour, but not the role player itself [35, 36]. Integrating instructional roles into the metadata model can solve the problem of annotating differ- ent theories and instructional principles in the learning de- sign [36]. In other words, instructional roles can facilitate learner modelling, enable automatic modelling, and are initi- ated as centres of reference. As illustrated in Figure 3, the proposed hierarchy of educa- tional learning objects aims to represent the different instruc- tional roles for learning resources. The hierarchy compo- nents were identified from the confirmed requirements de- sign features of the HWBM LS, which investigated in [19, 32]. Each class of the proposed metadata represents a partic- ular instructional role that allows mapping, exchange, reuse and search at this level. The proposed hierarchy presents a set of categories; and each instructional role identifies a set of vocabulary within a category. The Educational_object is the root of the metadata structure. Two main classes are identified as subclasses of Educational_object, i.e. the Fun- damental_concept and the Auxiliary_concept. Both classes are grouped into four categories of learning objects (i.e., theoretical, procedural, practical, and interactive) according to the confirmed learning content design features of HWBM LS. Fundamental_concept refers to the main learning objects being presented for the whole lesson (covers a number of topics) in a particular course. Auxiliary_concept covers the supplementary knowledge or resources being presented i.e. presentation of the details of each topic in a particular les- son. For instance, theoretical classes are subsumed under Fundamental_concept and Auxiliary_concept. Theoretical class for Fundamental_concept can be presented by a num- ber of learning objects such as objectives, prerequisites, problems and individual assignments. On the other hand, theoretical class for Auxiliary_concept can be presented by a number of resources and learning objects such as book chap- ters, flowcharts, and explanations. The aforementioned descriptors are structured based on the HWBM LS. However, a WBES can infer the actual learning preferences of learners by analysing the their behavioural interaction with the designed learning objects described by these metadata (e.g., time spent, hit rate and visited rate on each learning object). Furthermore, the hierarchy of educa- tional metadata is useful in gathering more behavioural in- formation since the information about the visited learning resources will be identified later by the designer or teacher. The proposed structure of the instructional role is frequently associated with the diverse population of learning objects that cover all requirements of learners according to the Mohammed Ahmed Ghazal, Nur Azan Mat Zin, Zurina Muda / Designing Domain Model For Adaptive Web-based Educational System According to Herrmann Whole Brain Model (2016) 70 brain-based structure. A teacher has to annotate these learn- ing objects (static descriptions) once only. The behavioural interactions of a learner with the WBES are used to annotate the dynamic descriptions. Therefore, learner modelling based on metadata is dependent on both static and dynamic descriptions. Fundamental_concept  Auxiliary_concept  Theoretical  Pre-requisites  Objective  FAQ  Wiki  Individu- al_assignment  Open_question  Theoretical  Reference  Flowchart  Explanations  Rule  Procedural  Guideline (In- structions)  Exercise  Brochures (Catalogue)  Wizard  Procedural  Slideshow  Tutorial  Notebook  Practical  Introduction  Video_tour  Group_assign ment  Group_discussi on  Practical  Example  Simulation (Try and error)  Case study  Interactive  Abstract  Overview  Outline  Mind_map  Summary  Multi- ple_choices  Comprehen- sive_exam  Interactive  Animation_flash  Flash_cards  Interactive_game IV CONCLUSION AND FUTURE WORK Learning Styles of HWBM is a new approach to be used for designing a domain model for AWBES. The domain model is a basic feature of designing a learning environment. It can be used for tracking, auditing and identifying learner prefer- ences, and thus adapt the behaviour of a system to match his/her preferences. This research overcomes the limitations of incorporating the educational and pedagogical theory in describing and organizing the learning content of education- al systems. The outcome of the research aimed to help the web developer for building learner model implicitly as well as for adapting the learning contents of WBES to match the learner preferences according to the brain structure of learn- ers. Further research should be conducted to explore the impacts of domain model on designing AWBES. REFERENCES [1] Anido, L.E., et al., Educational metadata and brokerage for learning resources. Computers & Education, 2002. 38(4): p. 351-374. [2] Committee, I.L.T.S., IEEE standard for Learning Object Metadata, in 1484.12.1. 2002. [3] Brooks, C. and G. McCalla, Towards flexible learning object metadata. Int. J. Cont. Engineering Education and Lifelong Learning, , 2006. Vol. 16, Nos. 1/2, 2006. [4] Friesen, N., Interoperability and learning objects: An overview of e-learning standardization. Interdisciplinary Journal of Knowledge and Learning Objects, 2005. 1(1): p. 23-31. [5] Intuitive, D.C.M., Dublin Core Metadata Element Set, in version 1.1. 2012. [6] Johnson, K. and T. Hall, Designing and Testing an Open- Source Learning Management System for Small-Scale Users in E-Learning Networked Environments and Archi- tectures, S. Pierre, Editor. 2007, Springer London. p. 209- 250. [7] Ullrich, C., The learning-resource-type is dead, long live the learning-resource-type. Learning Objects and Learn- ing Designs, 2005. 1(1): p. 7-15. [8] Gascueña, J.M., A. Fernandez-Caballero, and P. Gonzá- lez. Domain ontology for personalized e-learning in edu- cational systems. in Sixth IEEE International Conference on Advanced Learning Technologies (ICALT'06) 2006. IEEE Computer Society Press. [9] Ghazal, M.A., N.A.M. Zin, and Z. Muda, Towards Ap- plying a Brain-Based Learning Style Model to Improve Learner Modelling in an Adaptive Web-Based Educa- tional Systems (AWBES). International Journal of Digital Content Technology & its Applications, 2015. 9(2). [10] Ghazal, M.A., M.M. Yusof, and N.A.M. Zin. Adaptive Educational Hypermedia System using Cognitive style approach: Challenges and opportunities. in International Conference on Electrical Engineering and Informatics (ICEEI). 2011. Bandung: IEEE. [11] Inan, F., R. Flores, and M. Grant, Perspectives on the Design and Evaluation of Adaptive Web Based Learning Environments. Contemporary Educational Technology, 2010. 1(2): p. 148-159. [12] Lin, F., S. Graf, and R. McGreal, Adaptive and Intelligent Web-Based Educational Systems, in Guest Editorial Pref- ace. 2004, Retrieve from http://64.225.152.8/files/prefaces/ijwltt%20preface%204( 1).pdf [Last seen 23/7/2015]: Athabasca University, Can- ada. [13] Yang, T.-C., G.-J. Hwang, and S.J.-H. Yang, Develop- ment of an Adaptive Learning System with Multiple Per- Figure 3: Hierarchy of Educational Learning Objects Based on HWBM Educational_object http://64.225.152.8/files/prefaces/ijwltt%20preface%204(1).pdf http://64.225.152.8/files/prefaces/ijwltt%20preface%204(1).pdf Mohammed Ahmed Ghazal, Nur Azan Mat Zin, Zurina Muda / Designing Domain Model For Adaptive Web -based Educational System According to Herrmann Whole Brain Model (2016) 71 spectives based on Students' Learning Styles and Cogni- tive Styles. Educational Technology & Society, 2013. 16(4): p. 185-200. [14] Radwan, N., An Adaptive Learning Management System Based on Learner‟s Learning Style. International Arab Journal of e-Technology, 2014. 3(4): p. 7. [15] Lo, J.-J. and Y.-C. Chan, Design of Adaptive Web Inter- faces with Respect to Student Cognitive Styles, in Educa- tion and Educational Technology, Y. Wang, Editor. 2012, Springer Berlin Heidelberg. p. 331-338. [16] Chen, L.-H., Enhancement of student learning perfor- mance using personalized diagnosis and remedial learn- ing system. Computers & Education, 2011. 56(1): p. 289- 299. [17] Schiaffino, S. and A. Amandi, Intelligent user profiling, in Artificial Intelligence An International Perspective, M. Bramer, Editor. 2009, Springer Berlin Heidelberg. p. 193- 216. [18] Gena, C. and S. Weibelzahl, Usability Engineering for The Adaptive Web, in The Adaptive Web, P. Brusilovsky, A. Kobsa, and W. Nejdl, Editors. 2007, Springer Berlin / Heidelberg. p. 720-762. [19] Ghazal, M.A., N.A.M. Zin, and Z. Muda, Analysing the Relationship between Learner‟s Brain Dominance and Behavioural Learning Patterns of in Web-based Educa- tional Systems. Journal of Theoretical and Applied In- formation Technology, 2015. In Press. [20] Riding, R. and S. Rayner, Cognitive styles and learning strategies: Understanding style differences in learning and behaviour. 1998: D. Fulton Publishers. [21] Mampadi, F., et al., Design of adaptive hypermedia learn- ing systems: A cognitive style approach. Computers & Education, 2010. [22] Brown, E., et al., Evaluating learning style personaliza- tion in adaptive systems: Quantitative methods and ap- proaches. IEEE Transactions on Learning Technologies, 2009. January-March (2009 vol. 2)(1): p. 10-22. [23] Scales, P., Teaching in the Lifelong Learning Sector. 1st ed. 2008: Open University Press. [24] BECTA and N. Radwan. Learning styles : an introduction to the research literature. 2005 [cited 2014 8/6/2014]; Available from: http://dera.ioe.ac.uk/14118/1/learning_styles.pdf. [25] Vermunt, J.D., Metacognitive, cognitive and affective aspects of learning styles and strategies: A phenomeno- graphic analysis. Higher Education, 1996. 31(1): p. 25- 50. [26] Graf, S., Adaptivity in learning management systems focussing on learning styles, in Faculty of Informatics. 2007, Vienna University of Technology: Vienna. p. 192. [27] Coffield, F., et al., Learning styles and pedagogy in post- 16 learning: A systematic and critical review. 2004, Lon- don, England: The Learning and Skills Research Centre (LSRC). [28] Herrmann-Nehdi, A., The Best of Both Worlds - Making Blended Learning Really Work By Engaging the Whole Brain, in Better results through better thinking. Retrieved from http://www.hbdi.com/uploads/100016_whitepapers/1006 07.pdf [Last seen 22/7/2015]. 2009. [29] Bawaneh, A.K.A., et al., Jordanian Students' Thinking Style Based on Herrmann Whole Brain Model. Interna- tional Journal of Humanities & Social Science, 2011. 1(9): p. 89-97. [30] Lin, T. and Kinshuk, Cognitive Profiling in Life-Long Learning, in Encyclopedia of distance learning. 2009, In- formation Science Reference: Hershey, New York. p. 295. [31] Herrmann, N., The Theory Behind the HBDI® and Whole Brain® Technology, in Better results through bet- ter thinking. 1999, Herrmann International: Retrieved from http://www.hbdi.com/Home/?directory=100024_articles &actualFile=100543.pdf&saveName=Theory-Behind- The-HBDI.pdf [Last seen 22/7/2105]. [32] Ghazal, M.A., et al., Mapping Between Learning Style And Design Features Of Web-Based Educational Sys- tems. Under Publication, 2016. [33] Popescu, E., C. Badica, and P. Trigano, Description and Organization of Instructional Resources in an Adaptive Educational System Focused on Learning Styles, in Ad- vances in Intelligent and Distributed Computing, C. Bad- ica and M. Paprzycki, Editors. 2008, Springer Berlin / Heidelberg. p. 177-186. [34] Richter, C., W. Nejdl, and H. Allert. Learning Objects on the Semantic Web. Explicitly Modelling Instructional Theories and Paradigms. in Proceedings of World Con- ference on E-Learning in Corporate, Government, Healthcare, and Higher Education 2002. [35] Ullrich, C. Description of an instructional ontology and its application in web services for education. in Proceed- ings of Workshop on Applications of Semantic Web Technologies for E-learning, SW-EL. 2004. [36] Allert, H., H. Dhraief, and W. Nejdl. Meta-Level Catego- ry ‚Role‟in Metadata Standards for Learning: Instruction- al Roles and Instructional Qualities of Learning Objects. in The 2nd International Conference on Computational Semiotics for Games and New Media. 2002. University of Augsburg, Germany Mohammed Ahmed Ghazal. Assistant Professor at University College of Science and Technology (UCST). He has a PhD in In- formation technology from the National Universty of Malaysia in 2015 and a Master degree in Computer Science from Free Univer- sity Brussels in 2004. Currently, He is working as a head of re- search department at UCST. His research interests are adaptive web-based educational system, learner modelling, development of user interaction and usability applications, User-centered website design and development. Nor Azan Mat Zain: Associate Professor at the National Uni- versty of Malaysia. Currently, she is a head of Multimedia & Usa- bility research group. Her primary research interest is Advanced Technology for Learning. http://dera.ioe.ac.uk/14118/1/learning_styles.pdf http://www.hbdi.com/uploads/100016_whitepapers/100607.pdf http://www.hbdi.com/uploads/100016_whitepapers/100607.pdf http://www.hbdi.com/Home/?directory=100024_articles&actualFile=100543.pdf&saveName=Theory-Behind-The-HBDI.pdf http://www.hbdi.com/Home/?directory=100024_articles&actualFile=100543.pdf&saveName=Theory-Behind-The-HBDI.pdf http://www.hbdi.com/Home/?directory=100024_articles&actualFile=100543.pdf&saveName=Theory-Behind-The-HBDI.pdf Mohammed Ahmed Ghazal, Nur Azan Mat Zin, Zurina Muda / Designing Domain Model For Adaptive Web-based Educational System According to Herrmann Whole Brain Model (2016) 72 Zurina Muda: Associate Professor at the National Universty of Malaysia.. Currently, Sh is a member of Multimedia & Usability research group. Her primary research interests are Multimedia In- telligent Design and Development, Spatial Image Annotation and Retrieval, Interactive Game Design and Development.