The effectiveness of an induction algorithm for making decisions was investigated using a complete set of diagnoses and corresponding outcomes for all possible distinct cases of thyroid disorders. Investigation of both diagnoses and outcomes followed a late discovery of a misunderstanding between the knowledge engineers and the expert. The knowledge engineers had erroneously equated the expert's terms 'diagnosis' and 'outcome'. Following this discovery, two previously reported induced expert systems were re-evaluated using the more appropriate 'outcomes' and were found to be correct for 79 and 70% of all cases.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1016/s1386-5056(97)00060-9 | DOI Listing |
Publication Analysis
Top Keywords
knowledge engineers
12
misunderstanding knowledge
8
expert systems
8
engineers domain
4
domain experts
4
experts re-evaluation
4
re-evaluation induction
4
induction medical
4
medical expert
4
systems effectiveness
4
Similar Publications
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!