While FIG. 9 shows confusion matrix 900 for a binary classifier (i.e., a two classification system of “Yes” and “No” values), it is contemplated that the confusion matrix used for the disclosures herein can be extended to the case of more than two classes. For example, a confusion matrix may have three classes, e.g., high, medium, and low used to distinguish the accuracy of a predictive model based on the probability values associated with each of the images and where the image classifications are sorted in the confusion matrix into high, medium and low categories based on threshold values.
The confusion matrix 900 may be transmitted or stored as a data structure in a computing device, such as any of the computing devices descried for FIG. 1 or 2. In addition, while the confusion matrix 900 is shown in tabular format, the data structure of the confusion matrix may be presented in different data structure types, such as an array, a multi-dimensional array, a vector, a list, as hash, a dictionary, matrix, or other data structure.
