Objective
The objective of ISIDE was to develop an advanced model for integration of data- and knowledge-based management systems.
The ISIDE team worked on three main tasks:
-the specification of a knowledge representation formalism
-the specification of a logical inference formalism
-the definition of the architecture of an efficient storage and access layer using specialised VLSI hardware (Database machine).
The objective was to develop an advanced model for integration of database and knowledge based management systems. The project included the specification of a knowledge representation formalism, the specification of a logical inference formalism, and the definition of the architecture of an efficient storage and access layer using specialised very large scale integration (VLSI) hardware. From practical observations, it was concluded that no single universal paradigm would resolve the database programming problems. Instead, both a rule based approach and an object oriented approach, with capabilities to switch from one to the other, were advocated. Both approaches were intensively experimented upon, and prototype support systems demonstrated. An object server, GEODE, with sufficient flexibility to interface with rule based, object oriented and structured query (SQL) lanuguages, was built. On the architecture side, a consolidated draft was developed, emphasising future needs for large capacity core memories, to allow all the information needed during a transaction to be stored in the main memory. MALG, a theoretical algebraic machine, was built over GEODE.
From practical observations of the ARS/AGUSTA Helicopter Maintenance Training System (ITS) and CRIL's Software Engineering Documentation System (CACAO), ISIDE reached the conclusion that no single universal paradigm would resolve the database programming problems. Instead, both a rule-based approach (RDL1) and an object-oriented approach (ODL1/LDR2), with capabilities to switch from one to the other, were advocated.
Both approaches were intensively experimented upon, and prototype support systems demonstrated. The more mature, RDL1, already has a computational model (Predicate Transition Networks). An object server, GEODE, with sufficient flexibility to interface with rule-based, object-oriented and SQL languages, was built.
On the architecture side, a consolidated draft was developed, emphasising future needs for large-capacity core memories, to allow all the information needed during a transaction to be stored in the main memory. MALG, a theoretical algebraic machine, was built over GEODE.
Following performance evaluation studies for the different component systems, demonstrations were given of ITS with RDL1, Performance Evaluation, GEODE, and of ITS alone.
Exploitation
Some partners are looking into the possibility of developing a knowledge-based training product within the next two years, based on the RDL1 formalism.
The results of the ISIDE project constitute the basis forESPRIT II project STRETCH (number 2443).
Fields of science (EuroSciVoc)
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques. See: https://op.europa.eu/en/web/eu-vocabularies/euroscivoc.
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques. See: https://op.europa.eu/en/web/eu-vocabularies/euroscivoc.
- natural sciencescomputer and information sciencessoftware
- natural sciencescomputer and information sciencesknowledge engineering
- engineering and technologymechanical engineeringvehicle engineeringaerospace engineeringaircraftrotorcraft
- natural sciencescomputer and information sciencesdatabasesrelational databases
You need to log in or register to use this function
Programme(s)
Topic(s)
Data not availableCall for proposal
Data not availableFunding Scheme
Data not availableCoordinator
75512 Paris
France