Foundations on an adaptative tutoring system based on systemic networks

Abstract

In this paper, the elements which take part in a Tutoring System: knowledge organization, concepts, versions, evaluation, student typology and their activity, are considered as constituting a determined language: the teaching language.

This language can be formalized, according to its functional characteristics, context and interdependencies, by means of a systemic grammar.

It is possible to produce automatically a hierarchical graph, the systemic network, with the fore mentioned grammatical relations. The complete information required for a tutoring process can be extracted from this network through the Inference Machine, algorithm sets based on propagation and deduction rules.

Two objectives difficult to join in the tutoring systems, can be obtained by this way:

On one hand, to elaborate and consider, for the tutor or pedagogic team, the teaching elements and their relations in a straightforward way, independent of the generation and further extraction of network information.

On the other hand, a teaching-learning model which, without restricting the total student free activity, becomes fully adaptive to it, keeping stronger links to the tutor objectives and pedagogical criteria.