PeRES: A Personalized Recommendation Education System Based on Multi-agents & SCORM

Most E-Learning models proposed recently can offer personalized learning services for learners or make courseware reusable or portable. However, there are few models that can serve both purpose and none of them is designed to provide personalized services

  • PDF / 379,890 Bytes
  • 12 Pages / 430 x 660 pts Page_size
  • 47 Downloads / 151 Views

DOWNLOAD

REPORT


School of Computer, Wuhan University, PR China [email protected] [email protected] 2 Department of Computer Science, City University of Hong Kong, Hong Kong SAR, PR China [email protected] [email protected]

Abstract. Most E-Learning models proposed recently can offer personalized learning services for learners or make courseware reusable or portable. However, there are few models that can serve both purpose and none of them is designed to provide personalized services for both learners and instructors. This paper introduces an architecture of school-based personalized recommendation education system which can provide personalized services not only for diverse learners but also for instructors. In addition, it offers reusability and interoperability of courseware that is conformant with SCORM 2004 3rd Edition specifications. The architecture adopts multi-agents technology and consists of SIX software agents, which coordinate work hierarchy with each other to offer a range of primary functions that include static and dynamic users modeling, learning plan generation and adjustment, personalized content search, personalized recommendation, as well as real-time evaluation of learning progress. We provide the detail functional specification of these agents as well as a scenario walk-through of the architecture. Keywords: Personalized Learning, Recommendation System, Multi-Agents, SCORM.

1 Introduction With increasingly powerful hardware and advanced software technologies, it is possible for us to harness such technologies in education and, particularly, to cater to the diverse needs of learners as well as instructors. Riding on the successes of the deployment of web-based technologies in e-commerce, many researchers turn their focus to e-education. Kassim et al [15] has surveyed how the computer and Internet technology can play an important role in creating an effective web-based learning environment. While web-based applications in the two areas have very different goals and objectives, i.e. business vs. education, some of the core technologies in e-commerce make personalized e-education possible, particularly in areas such as personalized content search and recommendation, dynamic user modeling, learning plan generation and content annotation. H. Leung et al. (Eds.): ICWL 2007, LNCS 4823, pp. 31 – 42, 2008. © Springer-Verlag Berlin Heidelberg 2008

32

F. Zhu et al.

A few examples of Web-based intelligent learning environments have been illustrated in [17]. These previous works also highlight several problems in the WebBased Educational Environments: (i) Lack of a mechanism to interoperate with other learning systems and interchange information such as educational material, learning pattern and learning sequence etc; (ii) Model of the users may often to be updated due to influence such as users’ profile, users’ interests, and users’ activities; (iii) The key elements, which can be adopted to build and update the model of users from the view of designer of the learning systems, such as web pages visited b