Table 3.
Results of binary target classification. Two methods calculate various machine learning models’ performance measures: cross-validation performed on the data collected in the first part of the experiment and tested on a separate dataset of observations collected in the second part of the experiment.
CV on First Part Data | Test on Second Part Data | |
---|---|---|
Accuracy | 97% | 78% |
Recall of Pythium | 94% | 67% |
Recall of Phytophthora | 100% | 88% |
Precision of Pythium | 99% | 74% |
Precision of Phytophthora | 95% | 85% |