Home

myCBR is a project at DFKI to build a case based reasoning tool that can both be used as standalone application and as a Protégé plugin.

myCBR is open source, developed under the GPL license. Its aims are:

• to be easy to use,
• to enable fast prototyping,
• to be extendable and adaptable, and
• to integrate state-of-the-art CBR functionality.

Thus it supports the teaching and research of the CBR approach.

myCBR builds on top of Protégé, which already provides advanced functionality for defining and visualizing object-oriented case representations. However, myCBR-Retrieval will be usable as a standalone application as well.

News

• 09/06/30 New stable release: myCBR Version 2.6.4

Current Features

• Powerful GUIs for modeling knowledge-intensive similarity measures
• Similarity-based retrieval functionality
• Export of domain model (including similarity measures) in XML
• Additional stand-alone retrieval engine
• Extension to structured object-oriented case representations, including helpful taxonomy editors
• Powerful textual similarity modelling which distinguishes between word-based and character-based measures; it even supports regular expressions
• Scriptable similarity measures using Jython
• Rapid prototyping via CSV

Planned Features for myCBR version 3.0

• Integration of adaptation and completion rules
• Database connection
• Integration of similarity learning functionality
• Intelligent support for classification and regression tasks
• Improved scalability

Collaboration

jCOLIBRI is a generic CBR framework in Java developed by Universidad Complutense de Madrid – GAIA research group

There is a contribution which provides wrapper methods to use similarity measures generated with myCBR within jCOLIBRI. For more information, please visit http://gaia.fdi.ucm.es/projects/jcolibri/jcolibri2/contributions.html