Our fundamental scientific objective can be put forward by the following question: how far it is possible to go, in knowledge representation, by exclusively using graphs for representing knowledge and graph operations for representing reasonings? To study this question we use the classical four-stroke experimental methodology:
The basic formal model is the simple graphs model, which has three essential characteristics:
This basic model, which is derived from the simple conceptual graphs model of
Sowa, has been extended in several ways, keeping these fundamental properties
(cf. Mugnier's invited conference at ICCS'2000 for the more recent presentation of our approach).
Our current software tool is CoGITaNT (a descendant of CoGITo).
Development of graph-based Information Retrieval systems is the real problem we are presently involved in.