Related Experiment Videos
The use of machine learning program LERS-LB 2.5 in knowledge acquisition for expert system development in nursing
L Woolery1, J Grzymala-Busse, S Summers
1Childrens Mercy Hospital, MO.
Computers in Nursing
|November 1, 1991
Summary
Learning from Examples using Rough Sets Lower Boundaries (LERS-LB) can identify critical nursing data but requires accurate databases. Careful data preparation is essential for effective knowledge acquisition in expert systems.
Area of Science:
- Computer Science
- Artificial Intelligence
- Data Mining
Background:
- Expert systems require efficient knowledge acquisition.
- Rough set theory offers a framework for handling imprecise and incomplete data.
- LERS-LB is a program designed for knowledge acquisition using rough set theory.
Purpose of the Study:
- To evaluate the effectiveness of LERS-LB for knowledge acquisition from real-world data.
- To assess the suitability of different data formats (statistical files vs. databases) for LERS-LB.
- To explore LERS-LB's potential in identifying critical data for nursing practice.
Main Methods:
- Utilized LERS-LB (Learning from Examples using Rough Sets Lower Boundaries) for pattern extraction.
- Converted SPSS-X data files into decision-table format compatible with LERS-LB.
- Applied the 'dropping conditions' technique inherent in rough set theory.
Main Results:
- Both statistical files and databases can be converted to decision-table format.
- Well-developed databases are preferable to statistical files for LERS-LB processing.
- Data accuracy and completeness checks are crucial before running LERS-LB to prevent errors.
Conclusions:
- LERS-LB shows potential for identifying critical nursing data, reducing redundancy.
- Thorough database preparation is vital for successful knowledge acquisition.
- While LERS-LB aids in reducing the knowledge acquisition bottleneck, domain expert involvement remains essential for rule and system evaluation.