We test the classifier for different alpha thresholds. Recall that alpha is the threshold of the estimated probability, above which an observation is assigned to one category (positive class) and below to the other category (negative class).

Classification schemes keep evolving & improving with recent publications. Those recent techniques involve multi-output classifications, ie, the response variable/s is 2 or more in comparison to standard classification of just a single variable say Y. The multi-class MIMO SVR (multi input multi output - support vector regression) is one of those new techniques, eg: the multi output could be 3 variables (as Gender, Age-bracket, Earning-bracket) & may be denoted as [G, A, E], where gender is 2 class (male, female), age-bracket is multiclass (student, young-adult, adult, retired) & age-bracket is also multiclass. MIMO SVR can predict the 3 output variables class labels at once. The other multiclass MIMO schemes includes CANFIS (Co-Active Neuro-Fuzzy Inference System) & its variants.

In this fourth part of the tutorial we will discuss the ROC curve. The ROC curve is one of the methods for visualizing classification quality, which shows the dependency between TPR (True Positive Rate) and FPR (False Positive Rate). The more convex the curve, the better the classifier. In the example below, the „green" classifier is better in area 1, and the „red" classifier is better in area 2. AUC 1 means a perfect classifier, AUC 0.5 is obtained for purely random classifiers. AUC 0.5 means the classifier performs wor

In the last part of the tutorial we introduced quantitative indicators of classification model quality. In the next two parts we will take a closer look at a couple of graphical indicators. The first one is called the Confusion Matrix (the name „Contingency Table" is also used).