Transinformation standard

The standard transinformation R is not necessarily the maximum amount of information that can be obtained by a given classifier. The maximum amount R is called channel capacity C417, 4273. R would be obtained for a random sample of patterns with optimally adjusted a priori probabilities. Because the optimal a priori probabilities depend on the classifier, the channel capacity is not suitable for the evaluation of classifiers. The only useful starting point for the classification of unknown chemical patterns is to assume equal probabilities for all classes. 

The standard transinformation R is a function of the probabilities defined in Table 8. 

The probabilities in equation (136) can be computed from the predictive abilities and P2 by using equations (117) to (120). The determination of the standard transinformation from the predictive abilities for both classes is facilitated by the diagram shown in Figure 54. 

The concept of classifier evaluation by the standard transinformation suffers mainly from the complexity of equation (136) and from less intuitivness. On the other hand, the advantages are only a single number... 

FIGURE 54. Standard transinformation R for binary classifiers as a function of predictive abilities and P for both classes. Equal a priori probabilities for both classes are assumed C313D. 

---