Abstract | ||
---|---|---|
Experts can make highly accurate decisions, because they have accumulated a lot of (background) knowledge about the problem with theirs past experience. Because experience is very subjective different experts propose different diagnosis and decision based on the same facts gathered with observation of a patient. Machine learning methods also poses background knowledge encoded in theirs induction algorithms. In this paper we present a method for modifying this background knowledge and can therefore produce different hypothesis on same observation that therefore exposes different aspects e.g. opinions of experts. We also present a method for combining these hypotheses in combined, hopefully highly accurate, hypothesis by using boosting and multimethod approach. |
Year | DOI | Venue |
---|---|---|
2004 | 10.1109/CBMS.2004.1311720 | CBMS |
Keywords | Field | DocType |
decision support systems,decision trees,learning (artificial intelligence),medical diagnostic computing,medical expert systems,background knowledge,experts,machine learning methods,medical decision support,multiple opinions | Data science,Data mining,Decision tree,Intelligent decision support system,Computer science,Decision support system,Boosting (machine learning),Artificial intelligence,Clinical decision support system,Medical algorithm,Decision engineering,Machine learning | Conference |
ISSN | ISBN | Citations |
1063-7125 | 0-7695-2104-5 | 0 |
PageRank | References | Authors |
0.34 | 3 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
mitja lenic | 1 | 126 | 12.16 |
Petra Povalej | 2 | 24 | 5.70 |
Milan Zorman | 3 | 57 | 13.07 |
Peter Kokol | 4 | 309 | 74.52 |