| S. K. Bechhofer. The GRAIL classifier. Unpublished technical report, May 1995. Available from: University of Manchester, Medical Informatics Group, Department of Computer Science, Manchester M13 9PL, UK. |
....for time in the graph in figure 5.4. The other notable point about this data is that the query times show very little variation with increasing model size. The algorithm used by the GRAIL classifier gives a possible explanation for the large difference between the classification and query times [Bec95]. GRAIL concepts have at most one primitive base and the classifier optimises its search for subsumers by starting at the primitive base instead of at bottom. After finding all subsumers in this reduced search the classifier would normally have to search the rest of the model for any additional ....
S. K. Bechhofer. The GRAIL classifier. Unpublished technical report, May 1995. Available from: University of Manchester, Medical Informatics Group, Department of Computer Science, Manchester M13 9PL, UK.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC