Academic Open Internet Journal ISSN 1311-4360
|
Volume 23, 2008
|
Knowledge Representation in Personalized ELearning
H. Srimathi
Senior Lecturer, Department of Computer Applications, SRM University, Kattankulathur, 603 203
Dr. S.K. Srivatsa
Professor, St. Joseph’ College of Engineering, Chennai 119, India
Abstract
Adaptation is so natural, on the fly for teaching by humans, but it is a challenging issue of distance education. E-learners can experience the best learning when the web based material provides interactive communication; and information presented in a different ways with the control over learning. Knowledge about a user inferred from user interactions with the Elearning systems is used to adapt offered learning resources and guide a learner through them. Reasoning on the knowledge can be applied to adapt access, presentation, and navigation in the information resources. Semantic web technologies allow to link information of learning objects thus moving from document centric idea of current web to more fine grained semantic structures. The study is made on the derived Instructional Design Template for further incorporation with the SCORM model which has shifted the course delivery from simple presentation to learning objects. The knowledge items (learning objects) are linked to commonly agreed ontology. This enables construction of a user-specific course, by semantic querying for topics of interest. The paper highlights the work carried on designing architecture of Elearning portal with the personalization using ontology.
Keywords: Semantic web, Ontology, OWL, SOA, SCORM
1. INTRODUCTION
Effective ELearning thrives at the nexus of web usability, communication, relationship, document, and Knowledge Management tools. Compared with a human instructor, technology is less adaptive and once a plan of integration is implemented it is less likely to change according to student’s reactions. Educational servers should be created to possess enough intelligence to arrange for personalization of the learning tasks. In fact, from the learner’s perspective the server should appear to act as an intelligent tutor with both domain and pedagogical knowledge to conduct a learning session.
The key characteristic of Semantic Web architecture promises a powerful approach to satisfy the Elearning requirements. Learning material is semantically annotated. According to user preference, the learning material can be combined. The process is based on semantic querying and navigation through learning materials, enabled by the ontological background. This paper presents an effort toward to develop a semantic solution in Elearning system. This paper is organized as follow: Section 2 reviews related works done; section 3 describes the design followed by conclusion.
2. RELATED WORK
Interactive multimedia provides the stimulation for students to be actively involved in their learning. They can be encouraged to think, interact and gain a better understanding of the content material by providing user interaction and self-check exercises. Example of user response and the immediate feedback through self-check exercise are given in Fig 1. and Fig 2.
Fig. 1. Example of User interaction Fig. 2. Example of Immediate Feedback
Learning is a cognitive activity that differs from student to student. It has necessitated the design of effective Instructional Template on Learning Content Management System (LCMS). The revolution in the field of Information technology and evolving E-learning standards like SCORM has motivated further study on creating platform independent learning content. The Learning Management System (LMS) of SCORM can be applied to simple course management systems or highly complex enterprise-wide distributed environments (Fig. 3). It refers to a suite of functionalities designed to deliver, track, report on and manage learning content progress and learner interactions.
Fig. 3. A generalized model of LMS defined by SCORM
SCORM [1] focuses on interface points between content and LMS environments and is silent about the specific features and capabilities provided within a particular LMS and its components. The initial study is carried out to derive the model of LMS and the learning content model. Referring the Fig. 3, the detailed LMS is defined for the proposed Virtual Learning Environment (VLE). It is described in the Fig. 4. The required components of the LMS are Learning Content Management System (LCMS) for course delivery, Student Information System (SIS), which provides the interface to the individual users, Test Builder System (TBS) for building assignment and online examination, and Feedback analyzer system (FAS) for identifying the usability of interface and the course delivery. The Learning content model is derived by applying the Instructional Design on Educational and Information Technologies. The main objective of such mapping is to identify the best interactive environment on web based education [11, 12] and the factors’ which has direct or indirect influence on learning outcome. Fig. 5 is derived carefully to describe the generalized structure of “item” in the Learning Content Management System. The applicability of the identified components defined in LCMS is based on the course and learners’ preferences.
Fig. 4. The LMS for the proposed VLE Fig. 5. LCMS with LO hierarchy
3. DESIGN
The well-formed structure of Elearning with the multimedia design will be heightened when it is dynamically personalized to individual learners. The learners learning experience will be natural / “on the fly” based on their learning experience. The intelligence of a system includes the knowledge which has three dimensions; learning style, knowledge space and didactic resources and allows the system to provide adapted content fitting to the needs of each individual student [6]. The proposed framework to enable the individualized adaptation in an Elearning system is shown in Fig. 6. It pools to handle learner profiles, learning fragments, and adaptation rules as well as the adaptation engine responsible for providing personalized responses. The ontology based semantic rules can be formed on server to promote Adaptive education (see Fig. 7).
Fig. 6. Adaptive Learning Sequence Fig. 7. Educational Intelligent Server
3.1 Adaptation Engine
Domain database and profiles are used to build all formative paths. To generate a formative path, several approach are possible [5]. Here, paths finalized to the study of a specific matter are derived from the graph contained in the domain database to the engine component. The engine extracts from domain database all paths having the desired course unit as final node, and starting from initial knowledge [4, 8]. In Fig. 8, the example of paths retrieval for a given learner is shown. We suppose that the target course unit is the node 7 and that the student owns the knowledge provided by node set (2, 4), thus all paths for that student will start with node 2 or 7. After all paths are retrieving and eventually optimized, the engine filters them comparing arc weights with information stored on student profile.
3.2 SOA and Ontology
There are various vendors of LMS in the market that supports SCORM standard using its Run-time Environment (SCORM RTE) specification. The underlying architecture can also be developed with usage of .NET XML web services. The .NET Frameowork 2005 supports learner sequencing and personalization. The pedagogical intelligent agents are used to gather information for user according to his learning demand and matching learner profile. The suggested Service Oriented Architecture using web service and intelligent agent is formed in Fig. 9. It provides educational contents in the form of different services such as learning object repository, digital library, LMS, virtual class rooms, virtual labs, authoring services etc.,. The pedagogical agents can use SOAP, XML and HTTP to communicate over a network [9]. The Learning Objects are html pages, Flash SCOs (Shareable Content objects), Images, and etc.,
Fig. 8. Path retrieval for the given student Fig. 9. SOA ELearning using .NET
The sample XML representation of learner profile is shown in Fig. 10. This is the initial profile formed on first time login, and based on the query answered by the learner. It can be dynamically changed over a period of time; because of every action of learner is monitored and updated in learner profile. The first stage concerns the learning style of the learner, the second stage regards on skills of learner and the third stage highlights the type of learner. The adaptation mechanism of an Elearning system bases first on its observations retrieved by monitoring a learner’s interactions and second on his reactions to these observations. Therefore, the Elearning system should be flexible enough to provide the personalization. The XML is used to mere structuring the data. Therefore, OWL is used to provide semantic of XML information [7, 10]. The protégé software provides UI to promote ontology description. The sample ontology description is shown in Fig. 11.
Fig. 10. Initial Learner Profile Fig. 11. Ontology representation
4. CONCLUSIONS
This paper presents an approach for implementing the E-Learning scenario using semantic web technology. The instructional design principles are applied to identify the learning objects; LCMS and LMS. The Service Architecture is proposed for light-weight process. It is primarily based on ontology-based description of learning objects which promotes scalable, flexible and personalized access on Elearning. The learning objects are represented with high-degree relations by defining XML schema using ontology. The study is going to be enhanced with the complete implementation.
5. REFERENCES
[1] ADL (2004a). SCORM 2004 2nd edition - overview, retrieved March 22, 2006 from http://www.adlnet.org/downloads/70.cfm.
[2] Di Nitto, E., Mainetti, L., Monga, M., Sbattella, L. & Tedesco, R. (2006). Supporting Interoperability and Reusability of Learning Objects: The Virtual Campus Approach. Educational Technology & Society, 9 (2), 33-50.
[3] Olmedilla, D., Saito, N. & Simon. B. (2006). Interoperability of Educational Systems - Editorial of Special Issue. Educational Technology & Society, 9 (2), 1-3.
[4] Kassahun, A., Beulens, A. & Hartog, R. (2006). Providing Author-Defined State Data Storage to Learning Objects, Educational Technology & Society 9 (2), 19-32.
[5] Jesshope, C. Heinrich, E. & Kinshuk (2000) Technology Integrated Learning Environments for Education at a Distance. Paper presented at the DEANZ 2000 Conference, 26-29 April 2000, Dunedin, New Zealand.
[6] Wenying Guo, and Deren Chen (2006), Semantic Approach for e-learning System , (IMSCCS’06)
[7] Liangbin Wang, Guojin Zhu (2004), Developing an XML Schema of BWW ontologies for Semantic Web (ISPAN’04)
[8] Marko Rosic, Slavomir Stankov and Vlado Glavink (2002), Application of Semantic Web and Personal Agents in Distance Education, IEEE MELECON, Egypt,
[9] Konstantinos Votis, Christos, Alexakos, Kostas Giotopoulos and Spiridon, Likothansassis (2006), An Onoltogical Service Oriented and Web Searching mechanism in distributed e-learning Repositories, (ICALT’07)
[10] Paulo Gomes, Bruno Antunes, Luis, Rdorigues and et. al. (2006) Using Ontologies for elearning Personalization , 3rd Learning Conference, Portual,
[11] Timothy Brook Hall Ulim-EMRC, (2002) Curriculum Development Critical Success Factors for eLearning Implementation, University of Limerick, Ireland Proceedings of the International Conference on Computers in Education (ICCE’02)
[12] Siew-Woei Ling Chee-Weng Khong,, Chien-Sing Lee, (2001) An evolving instructional design model for designing Web-based courses, Educational Technology & Society
Technical College - Bourgas,
All rights reserved, © March, 2000