[Pellet-users] SWRL rules and consistency checking times

[Crapo, Andrew (GE, Research)]

I have a set of about 350 SWRL rules. When I call isConsistent() or
realize() (which appears to call isConsistent()), it never comes back. I
have tried it with different numbers of rules to try to understand what
might be happening. The underlying ontology (a simplified version used
for testing) is AL(D), 114 classes, 5 properties, and 3 individuals. The
strategy is ContinuousRulesStrategy.

Here's my data showing a sharp increase in time to check consistency:

Expressivity: AL(D), Classes: 114 Properties: 5 Individuals: 3 Strategy:
ContinuousRulesStrategy

Times are for Consistency, Expressivity, and Realization (in ms).
With tracing at debug:
	  1-rule: 63/0/1375
	 10-rule: 94/0/1390
	100-rule: 9031/0/1219

With tracing off:
	  1-rule: 156/0/141 (15/0/125, 16/0/140, 15/0/94)
	 10-rule: 109/0/110 (47/0/78, 47/0/125, 47/0/78)
	100-rule: 8969/0/94 (9031/0/110, 8860/0/15, 9000/0/31)

1) Why is the time to do consistency checking going up so rapidly? Is
there something I need to set differently?
2) Is it possible to not do consistency checking at all once in the
environment where the rule set gets realized repeatedly? 

Andy

Andrew Crapo
GE Global Research
Information Scientist

T 518 387 5729
F 518 387 6104
D *833 5729
E crapo at research.ge.com
www.research.ge.com

One Research Circle
Niskayuna, NY 12309 USA
General Electric Company

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.owldl.com/pipermail/pellet-users/attachments/20080331/f8c04906/attachment.htm