A computational model for developing semantic web-based educational systems

https://doi.org/10.1016/j.knosys.2009.02.012Get rights and content

Abstract

Recently, some initiatives to start the so-called semantic web-based educational systems (SWBES) have emerged in the field of artificial intelligence in education (AIED). The main idea is to incorporate semantic web resources to the design of AIED systems aiming to update their architectures to provide more adaptability, robustness and richer learning environments. However, the construction of such systems is highly complex and faces several challenges in terms of software engineering and artificial intelligence aspects. This paper presents a computational model for developing SWBES focusing on the problem of how to make the development easier and more useful for both developers and authors. In order to illustrate the features of the proposed model, a case study is presented. Furthermore, a discussion about the results regarding the computational model construction is available.

Introduction

Current artificial intelligence in education (AIED) systems have tried to incorporate semantic web resources to their design and architecture. The main idea behind this purpose is the attempt to represent information on the Web so that computers can understand and manipulate it, leading to more adaptable, personalized and intelligent learning environments. In fact, there is a significant interest within the AIED community on the evolution of e-learning systems in this direction. The new generation of AIED systems comes from the combination of two broad modalities of web-based educational systems: (i) e-learning systems (or learning management systems – LMS), which provide interaction between students and teachers through the use of information technology (by using synchronous and asynchronous tools) to ensure this communication, and (ii) AIED systems, which use artificial intelligence techniques to provide personalized interactions, aiming to improve the learning and problem solving processes. The result of such combination are the so-called semantic web-based educational systems (SWBES).

The construction of SWBES, however, is a rather complex task which faces challenges in terms of software engineering and artificial intelligence aspects, such as: extensibility, interoperability, contextualization and consistence of metadata, dynamic sequence of learning and contents, integration and reuse of content and artificial intelligence techniques, distribution of services and new models of learning [15]. Such issues have been influenced by the aim of representing information on the Web in a way computers can understand and manipulate it. Therefore, SWBES are assumed as the new generation of Web-based educational systems that uses semantic web technologies to generate more personalized, adaptable and intelligent educational systems [13].

This paper presents a computational model for the development of SWBES focusing on the problem of how to make the development easier and more useful for both developers and authors. Before, a reference model is introduced in order to be used in the definition of the computational model. In order to illustrate the features of the proposed model, a case study is presented. Furthermore, a discussion about the results regarding the computational model construction is available.

The remainder of this paper is structured as follows. An overview about the evolution of intelligent educational systems and some problems for building them are presented in Section 2. Section 3 presents a reference model adopted as the next generation of educational systems, here assumed as the semantic web-based educational systems. Section 4 presents a computational model for developing SWBES. A case study to illustrate the computational model is presented in Section 5. Section 6 presents a discussion about the results regarding the computational model construction. Related Work on SWBES is detailed in Section 7. Conclusions and future work are presented in the final section.

Section snippets

Overview of educational systems and research problems

This section provides a review of the computer-based educational systems, aiming to present sufficient background and open issues to understand the proposed model. It starts with a discussion about classical approaches to these systems before embarking on Adaptive e-learning environments and problems for building them.

Reference model

The semantic web (SW) extends the classical web in the sense that it allows a semantic structure of web pages, giving support to humans as well as artificial agents to understand the content inside the web applications. As a result, Semantic Web provides an environment that allows software agents to navigate through web documents and execute sophisticated tasks. SW itself offers numerous improvements in the context of Web-based educational systems contributing to the upgrade of learning quality.

Computational model

This section describes a computational model, called e-Mathema, for building semantic web-based educational systems based on the proposed reference model. Some applications have been created by using e-Mathema, such as [9], [7], [10], [5]. This model has three layers which are presented in Fig. 2 followed by the description of each layer.

The architecture was developed as a multi-layer architecture whose layers are described below [14]:

  • Framework: it is maintained by developers who can add new

Case study

The aim of this section is to illustrate the features of the proposed computational model through the development of an intelligent tutoring system.

The system is applied at legal domain, providing Law students with real cases, rules and different viewpoints of a given body of knowledge. The main idea is to engage Law students into interactions with the system based on the resolution of Legal problems and their consequences on other tutorial activities, concerning the penal Law. The main

Discussion of the addressed problems

This Section aims at discussing how each problem stated in Section 2.3 was addressed by the proposed computational model. During the system implementation and evaluation in a real scenario, the solutions adopted to these problems are discussed as follows.

  • High development cost: it was decreased through the reuse of an agent-based architecture, ontologies, and a methodology that specifies the main aspects of a SWBES;

  • Complexity to develop AI algorithms: it was solved through the reuse of some

Related work

Some tools for building educational systems have been created. A relevant analysis of the state of the art can be viewed in [23] in a traditional perspective and other proposals were developed under the perspective of semantic web technologies [2], [31], [33]. However, recently, some new environments have been developed. Some of them, considering the proposals closely related to the presented proposal here are described below.

Millard [29] proposes the use of semantic web services for building

Conclusions

In this paper a computational model for developing semantic web-based educational systems was described. The approach introduced in this work promotes an easy and efficient way to build such systems for both authors and developers. This computational model is characterized by offering low development costs, scalability, extensibility, interoperability, and low maintenance costs. Moreover, with this approach it is also possible to deal with the development of artificial intelligence, interactive

References (37)

  • T. Anderson et al.

    The educational semantic web: visioning and practicing the future of education

    Journal of Interactive Media in Education – JIME

    (2004)
  • V. Aleven et al.

    The cognitive tutor authoring tools (CTAT): preliminary evaluation of efficiency gains

  • L. Aroyo, A. Inaba, L. Soldatova, R. Mizoguchi, Ease: evolutional authoring support environment, in: Intelligent...
  • L. Aroyo et al.

    The new challenges for e-learning: the educational semantic web

    Educational Technology and Society

    (2004)
  • C. Bezerra, I.I. Bittencourt, E.B. Costa, A semantic web-based tool for generating pedagogical evaluations (in...
  • I.I. Bittencourt et al.

    Towards a new generation of web-based educational systems: the convergence between artificial and human agents

    IEEE Multidisciplinary Engineering Education Magazine

    (2008)
  • I.I. Bittencourt, M. Tadeu, E.B. Costa, R.N. Oliveira, A.P. Silva, Combining AI techniques into a legal agent-based...
  • I.I. Bittencourt, C. Bezerra, C. Nunes, E. Costa, R.N. de Oliveira, M. Tadeu, A.P. da Silva, Ontologia para construção...
  • I.I. Bittencourt, B. Fonseca, G. Maia, J.S.S. Melo, R. Pinho, E. Ferneda, E.B. Costa, A.P.B. Silva, L.M. Brasil, An...
  • I.I. Bittencourt, E.B. Costa, B. Fonseca, G. Maia, I. Calado, Themis, a legal agent-based ITS in AIED 2007 workshop on...
  • I.I. Bittencourt, E.B. Costa, H.O. Almeida, B. Fonseca, G. Maia, I. Calado, A.P. Silva, Towards an ontology-based...
  • I. Bittencourt et al.

    An authoring system for building agent-based interactive learning environments (in Portuguese)

    Revista Brasileira de Informática na Educação

    (2007)
  • I.I. Bittencourt et al.

    Researchs directions on semantic web and education

    Scientia Journal

    (2008)
  • I.I. Bittencourt, E.B. Costa, B. Fonseca, G. Maia, J.S.S. Melo, E. Ferned, A.P.B. Silva, Constructing intelligent...
  • C. Brooks, J. Greer, E. Melis, C. Ullrich, Combining ITS and elearning technologies: opportunities and challenges, in:...
  • W. Chen et al.

    Learner model ontology and leaner model agent

    Cognitive Support for Learning – Imagining the Unknown

    (2004)
  • V. Chepegin, User Modelling....
  • E. Costa et al.

    From a tridimensional view of domain knowledge to multi-agents tutoring systems

  • Cited by (82)

    • How can artificial intelligence impact sustainability: A systematic literature review

      2022, Journal of Cleaner Production
      Citation Excerpt :

      Intelligence in AI is an ability to do reasoning, problem solving, learning, and integrate various human functions like perception, attention, memory, language, or planning. Reasoning in AI helps machines think rationally and apply deductive, inductive, abductive, and monotonic approaches to available sets of data, facts, and knowledge to make constructive predictions and valid conclusions (Bittencourt et al., 2009). Another aspect of AI is learning, which improvises the knowledge of an AI program and is classified into supervised, semi supervised, unsupervised, and reinforcement learning.

    • Personalized assessment model for alphabets learning with learning objects in e-learning environment for dyslexia

      2020, Journal of King Saud University - Computer and Information Sciences
      Citation Excerpt :

      Data properties: It associates the resource to some literal values. Semantic Rules: Semantic Web Rule Language (SWRL) is developed using description logic (DL) and HORN logic (Bittencourt et al., 2009). SWRL is a combination of work from DARPA agent markup language and RuleML (Rani et al., 2015).

    • An object triple mapping system supporting detached objects: A performance and memory usage empirical comparison

      2017, Engineering Applications of Artificial Intelligence
      Citation Excerpt :

      For instance, e-commerce applications use ontologies for parametric searches, navigation improvements, and heterogeneous systems integration (Cruz, 2002). Approaches for building educational systems have also been using ontologies to represent knowledge about students, educational resources and pedagogical strategies as well as for providing interoperability between these systems (Bittencourt et al., 2009). More recently, developers working with mobile devices launched virtual assistants based on ontologies (Das et al., ).

    View all citing articles on Scopus
    1

    Tel.: +55 83 3310 1124, fax: +55 83 3310 1273.

    View full text