Effects of Pooling Samples on the Performance of Classification Algorithms: A Comparative Study
Figure 4
Performance of classifiers based on the animal scenario. Performance of classifiers for datasets of individual samples and pool sizes of 2, 3, and 5 are compared. Misclassification rates rise with larger pool sizes as the human scenario. (a) and (b) show the comparison when numbers of markers are 2 and 10, respectively. The height of bars indicates 95% confidence interval from 300 replications.
