IMCS/Publications/CSJM/Issues/CSJM v.12, n.1 (34), 2004/

Symbolic Deductive Reasoning using Connectionist Models

Authors: Mihai Horia Zaharia, Florin Leon, Dan Galea
Keywords: Artificial intelligence, cognitive agents, schemata, knowledge representation, deductive reasoning, neural networks.


In this paper, we try to combine the possibility of symbolic deductive reasoning with the learning capability of the connectionist models. We introduce several algorithms for learning relations between concepts and finding paths in a transitive manner between learned concepts. An application that implements the proposed model is described and a number of case studies are presented.

Mihai Horia Zaharia, Florin Leon, Dan Galea
Department of Automatic Control
and Computer Engineering
Technical University "Gh. Asachi"


Adobe PDF document0.40 Mb