Related Experiment Videos
A rational reconstruction of INTERNIST-I using PROTEGE-II
M A Musen1, J H Gennari, W W Wong
1Section on Medical Informatics, Stanford University School of Medicine, CA 94305-5479, USA.
Proceedings. Symposium on Computer Applications in Medical Care
|January 1, 1995
Summary
PROTEGE-II facilitates principled development of knowledge-based systems. This methodology, applied to INTERNIST-I, showcases domain ontologies and reusable methods for creating functional programs.
Area of Science:
- Artificial Intelligence
- Knowledge Engineering
- Software Engineering
Background:
- Knowledge-based systems require principled development and maintenance methodologies.
- Existing tools may lack flexibility for complex system reconstruction.
- The INTERNIST-I system is a well-established medical expert system.
Purpose of the Study:
- To evaluate the PROTEGE-II methodology for building and maintaining knowledge-based systems.
- To demonstrate the application of PROTEGE-II in reconstructing the INTERNIST-I system.
- To highlight the utility of domain ontologies, reusable methods, and declarative mappings.
Main Methods:
- Utilized the PROTEGE-II methodology and its suite of tools.
- Reconstructed the INTERNIST-I knowledge-based system.
- Employed domain ontologies, reusable problem-solving methods, and declarative mapping relations.
Main Results:
- Successfully reconstructed the INTERNIST-I system using PROTEGE-II.
- Demonstrated the effectiveness of PROTEGE-II components in program creation.
- Generated a domain-specific knowledge-acquisition tool with functionality similar to QMR-KAT.
Conclusions:
- PROTEGE-II provides a principled approach to knowledge-based system development.
- The study enhances understanding of the PROTEGE-II methodology and INTERNIST-I's underlying models.
- Domain ontologies and reusable methods are crucial for efficient system reconstruction.