Factsheet: Case Base Reasoning

Definition

Case-Based Reasoning is:

A methodology for modelling of human cognition.
A methodology of artificial intelligence (AI) used for electronic encapsulation and reuse of knowledge.
A methodology for development of intelligent computer systems.

Application objectives

Use of experience (reuse of knowledge) in order to solve a problem and`create a new experience (a new piece of knowledge).

Pertinent participation process phase(s)

	1.	Starting organization
	2.	Actors analysis, context
	3.	Diagnostic of the current situation
	4.	Search of solutions
	5.	Implementation, evaluation

Application method

Nearest- Neighbour method often combined with Dynamical Induction`.

Application example(s)

CBR for in situ soil remediation (N. Kukuric, TNO) The references contain extensive overview of developed applications – see below.

Example tools

CBR Works 4 – this programme has been frequently used in the past but is no longer marketed by Empolis http://www.empolis.com/. Upon request it is available from the Empolis FTP servers for research. Adaptability: very limited (unless the source-code is provided).

Ownership: Empolis http://www.empolis.com/

Licensing: a research license available (not for commercial use). There are many tools available in open source, most of which are university prototypes.

A list of links to Case-Based Reasoning tool vendors and consultants is available at: http://www.ai-cbr.org/tools.html

AquaStress contact(s)

Neno Kukuric, TNO (neno.kukuric@tno.nl)

Reference

Agnar Aamodt, Enric Plaza, 1994, Case-Based Reasoning:Foundational Issues, Methodological Variations, and System Approaches, AICom - Artificial Intelligence Communications, IOS Press, Vol. 7: 1, pp. 39-59