Elsevier

Information Sciences

Volume 88, Issues 1–4, January 1996, Pages 275-298
Information Sciences

Intelligent system
Neural networks applied to knowledge acquisition in the student model

https://doi.org/10.1016/0020-0255(95)00233-2Get rights and content

Abstract

Knowledge acquisition for the student model of intelligent tutoring systems (ITSs) remains a difficult problem, partly because of the complexity associated with understanding both how people learn and how it is best to tutor, much of which relates to metacognition and problem-solving skills. The bottleneck associated with this area significantly increases the development times of ITSs. Neural networks have made a marked impact in many artificial intelligence areas such as pattern recognition, speech learning, speech understanding, and hand-written character recognition. Neural networks are noted for their ability to handle noise and approximate data, to generalize over situations they have not handled before, and to be represented in a way amenable to parallel processing. In addition, they have the ability to learn, a characteristic which should prove very useful in the development of ITSs. In this paper, we show that neural networks can address the knowledge acquisition bottleneck associated with the student model. We demonstrate that incomplete knowledge obtained from the expert can be refined and expanded by a neural network to provide a more complete, and hence more accurate, student model.

References (27)

  • J. Brown et al.

    Diagnostic models for procedural bugs in basic mathematical skill

    Cog. Sci.

    (1978)
  • L. Fu et al.

    Mapping rule-based systems into neural architecture

    KnowledgeBased Syst.

    (1990)
  • M. Caudill

    Using neural nets: Hybrid expert networks

    AI Expert

    (1990)
  • S. Derry et al.

    Local cognitive modeling of problem-solving behavior: An application of fuzzy theory

  • S. Derry et al.

    Characterizing the problem solver: A system for on-line error detection

  • L. Hawkes et al.

    Error diagnosis and fuzzy reasoning techniques for intelligent tutoring systems

    J. Artificial Intell. Education

    (1990)
  • D. Holmes

    An Informal Reasoning Technique And Truth Maintenance Subsystem For Global Diagnosis In An Instructional System

  • J. Hopfield et al.

    Computing with neural circuits: A model

    Science

    (1986)
  • S. Katz et al.

    Modelling the student in SHERLOCK II

    J. AI in Education

    (1992)
  • D. McArthur et al.

    Tutoring techniques in algebra

    Cog. Instruc.

    (1990)
  • S. Mengel et al.

    Using a neural network to predict student responses

  • S. Mengel et al.

    On the use of neural networks in intelligent tutoring systems

    J. AI in Education

    (1991)
  • Cited by (12)

    • Knowledge modeling via contextualized representations for LSTM-based personalized exercise recommendation

      2020, Information Sciences
      Citation Excerpt :

      This neglects the mutual effect between learning processes on these parameters by assuming each knowledge acquisition process is independent [44]. Researchers have successfully demonstrated that neural networks are effective at modeling students, at facilitating the acquisition of knowledge, and at identifying relations between skills [29,31]. Although the approach used in the cited papers is dated in terms of data availability and computing resources, they nevertheless show the effectiveness of using layered neurons and adjusted weights in modeling complex human activities such as learning.

    • Monitoring students' actions and using teachers' expertise in implementing and evaluating the neural network-based fuzzy diagnostic model

      2007, Expert Systems with Applications
      Citation Excerpt :

      A comprehensive review can be found in Jameson (1996). Neural networks have also been proposed in student modeling due to their abilities to learn from noisy or incomplete patterns of students’ behavior and generalize over similar examples (Beck & Woolf, 1998; Posey & Hawkes, 1996). This generalized knowledge can then be used to recognize unknown sequences.

    • Neuro-fuzzy knowledge processing in intelligent learning environments for improved student diagnosis

      2005, Information Sciences
      Citation Excerpt :

      However, several attempts have been made to incorporate the powerful learning abilities of neural networks in existing student modelling systems taking advantage of synergies with other AI methods. A hybrid approach, where each node and connection has symbolic meaning, has been proposed in TAPS [46]. The back-propagation algorithm has been used to modify weights that represent importance measures of attributes associated with student's performance, in order to refine and expand incomplete expert knowledge.

    View all citing articles on Scopus
    View full text