The University of Edinburgh -
Division of Informatics
Forrest Hill & 80 South Bridge

MSc Thesis #9611

Title:Knowledge Inference Engine (Kie)
Date: 1996
Abstract:How do we avoid the problem of the exponential increase in calculation-time due to the increase in the complexity of calculation with the increasing size of a knowledge base ? This paper proposes a solution illustrated by a knowledge base which has been implemented using frames to test the possibility of ignoring the irrelevant and just taking relevant information into account. Relevant information is found through the use of a proximity measure which is based on the minimum number of links that it takes to traverse from one piece of information to another. The domain of this knowledge-base is scene-interpretation and this paper therefore also offers an approach to describing behaviour in this context.

[Search These Pages] [DAI Home Page] [Comment]