Naive Bayes Classifier Results - Posterior Tab
Naive Bayes
Select the Posterior tab of the Naive Bayes Classifier Results dialog to access options to generate a Posterior graph given your choice of independent variables. This is useful for observing the effect of adjusting an independent variable on the Posterior distributions (and, hence, predictions). This type of graph falls into a wider category known as Response graphs. Since the graph is two dimensional, all the independent variables must be fixed and one be selected as the x-axis variable with a specified range. Thus, while one specially chosen (see below) independent variable is altered (and plotted across the x-axis of the graph), values must also be provided for the other independent variables of the Naive Bayes Classifier. These are given fixed values, and are referred to as fixed independents (see below). Thus, the Posterior graph actually represents a one-dimensional slice through an N dimensional Posterior surface, where N is the number of independent variables.
Fixed independent list. This spreadsheet displays the values of the fixed independent variables.