Paper Info
Title
Where is Knowledge in Computational Intelligence?: On the Reduction of the Knowledge Level to the Level Below
Abstract
The fundamental conjecture in the present work is that it is actually possible to find effective procedures for reducing the knowledge level to the level below (from natural language to code) and vise-versa, if the knowledge is structured. In order to support this conjecture, we shall prove that this procedure exists and is reversible at least for the case of hierarchical classification and its associated problem solving method (PSM) ("established" and "refined"), according to Chandrasekaran's proposal and for neural networks.As conclusions we shall emphasize two aspects, (1) the need of clarifying "where is the knowledge" in regard with any computation, and (2), our conviction that most of the work must be done by modelling knowledge on tasks and PSM's at the knowledge level and within the external observer's domain, there is where most of the knowledge remains, either it is said explicitly or not.
Year
DOI
Venue
1998
10.1109/EURMIC.1998.708094
EUROMICRO
Keywords
Field
DocType
fundamental conjecture,computational intelligence,effective procedure,knowledge level,neural network,modelling knowledge,present work,natural language,hierarchical classification,external observer,associated problem,neural networks,artificial intelligence,natural languages,neural nets,ontologies,computer languages,fuzzy systems,knowledge based systems
Procedural knowledge,Commonsense knowledge,Symbolic artificial intelligence,Knowledge level,Computer science,AI-complete,Mathematical knowledge management,Knowledge-based systems,Artificial intelligence,Artificial Intelligence System
Conference
Volume
ISSN
ISBN
2
1089-6503
0-8186-8646-4-2
Citations 
PageRank 
References 
3
0.51
7
Authors
3
Name
Order
Citations
PageRank
José Mira154371.44
J. C. Herrero250.97
A. E. Delgado330.51