Purpose Based Learner Modelling

This paper presents a brief review of early learner modelling in Intelligent Tutoring Systems focusing particularly on procedural knowledge representations vs. declarative knowledge representations. It then tracks the paradigm shift from traditional learner modelling, which emphasises knowledge representation, to distributed learner modelling, which focuses on the modelling process. Learner models in distributed, multi-agent environments are created by different agents for different purposes, are often fragmented and local, and only make sense in the specific context of creation. My work, in particular, is aimed at defining a taxonomy of purposes and retrieving information relevant to a particular purpose “just-in-time” in order to assemble and integrate fragmented learner model information about a learner. Examples of my approach will be drawn from the I-Help system. Some open issues in this approach are also given in this paper, which motivate my further investigation and exploration.