Abstract
Diagnosis is characterized as an exercise in classification,
where the task is to assign a crop to a risk group as a
basis for evidence-based crop protection decision making.
Underlying the process of diagnostic decision making is
Bayesian updating of probabilities. Alongside updating of
probabilities, assessments of diagnostic information allow
further description of the characteristics of diagnostic tests,
and of the predictions made on the basis of test outcomes. This
is illustrated analytically, graphically (by means of isoinformation
contour plots and information graphs) and by
discussion of example epidemiological scenarios.
Original language | English |
---|---|
Pages (from-to) | 267 - 286 |
Number of pages | 20 |
Journal | Australian Plant Pathology |
Volume | 43 |
Issue number | 3 |
DOIs | |
Publication status | First published - 2014 |
Bibliographical note
1023370Keywords
- Bayes' theorem
- Entropy
- Expected mutual information
- Relative entropy
- Specific information
- Weight of evidence