[Pellet-users] Reasoning Efficiency in Pellet

[Bijan Parsia]

On Tue, September 4, 2007 12:25 pm, anushya marcus <anushyamarcus at yahoo.co.uk> said:

>   Hello,
> 
>   Are there any references that gives an indication of the efficiency of the ABox
> and TBox reasoning in Pellet.

It really varies enormously. I don't think it's really possible to state, baldly, which is "better" than the other.
 
>   In the domain I am interested in, the entities could be modelled either as
> individuals or concepts (with certain restrictions). So I am wondering which
> option to choose, depending on what is more efficient, ABox or TBox reasoning.

That's just two little information to base anything on. You might look at the Suduko encoding (in Swoop's Advanced menu) which gives two encodings (one based on individuals and one based on classes, IIRC).

>   Any comments on this will be greatly appreciated.

I think that the basic rule is that the complexity of your problem determines all else and that experimentation is the only way forward.

Sorry.

Cheers,
Bijan.