[foaf-dev] Stereotype modelling

Hi,

I struggled recently with adaptable stereotype modelling. That means, I 
would like to use predefined stereotype settings to initialize a user 
profile (as also proposed here [1]). However, after that initialisation 
process, the user should be able to refine these settings to a certain 
(this should also depend on her/his "computer skills"). Furthermore, it 
should later also be possible to create stereotypes (clusters) on the 
user behaviour through user profile analysis.
Currently, I found the property foaf:membershipClass[2], which might be 
userful for that use case. Because, initially I also thought about 
creating sub classes of foaf:Person, which consists of some 
owl:Restriction instances. Furthermore, it might then be useful to 
connect these stereotype class definitions to some SKOS concepts. The 
concepts of the specific SKOS taxonomy can later probably be used, when 
the user adapted the user profile setting. However, one might to keep 
somehow the relation to initial stereotyp (something like: "this profile 
is still similar to a certain degree to stereotype X").
Furthermore, do you think, that it is necessary to introduce a new class 
for stereotypes on general?
Generally, one can create for every stereotype category different 
taxonomies, that means, e.g. several for "music listening type", several 
for "music knowledge type" and several for "computer skills type". For 
my project, I would then apply an exemplified taxonomy for each 
stereotype category. However, this stereotype category taxonomy mustn't 
be the optimal one.

Any thoughts?

Cheers,


Bob

[1] http://www.proceedings2005.imcsit.org/docs/93.pdf
[2] http://xmlns.com/foaf/spec/#term_membershipClass