Information Technology Reference
In-Depth Information
Fig. 3.3 A fuzzy-weighted qualitative overlay model [the fuzzy set in each node corresponds to
the qualitative values ('unknown', 'insufficiently known', 'known', learned')]
concepts are learned, which domain concepts are partly known and which domain
concepts are completely unknown.
3.3.3 Stereotypes
Another common used approach of student modeling is stereotyping. Stereotypes
were introduced to user modeling by Rich (1979) in the system called GRUNDY.
The main idea of stereotyping is to create groups of students with common charac-
teristics. Such groups are called stereotypes. In other words, a stereotype normally
contains the common knowledge about a group of users. A new user will be assigned
into a related stereotype if some of his/her characteristics match the ones contained
in the stereotype. Each stereotypes has a set of trigger conditions, which activate the
stereotype if they are true, and a set of retraction conditions, which deactivate the
stereotype if they are true to Kay (2000).
The stereotype student model of the presented hybrid student model is three-
dimensional (Fig. 3.4 ). The first dimension (KL) consists of stereotypes that rep-
resent the learner's knowledge level. They vary from novices to experts. The value
of KL is defined considering the information of the fuzzy-weighted qualitative
overlay model. A learner is classified to a knowledge level (KL) stereotype cat-
egory according to which domain concepts the learner knows and how well. The
particular stereotype category gives information about the learning material that
should be delivered to the learner. The second dimension (ErrTyp) consists of two
stereotypes and concerns the type of errors that a learner can make. It helps the
system to reason the learner's performance. For example, the system can infer if
Search WWH ::




Custom Search