Figure 11
The scatter plot illustrates the relationship between the number of added training samples on the x-axis and the value of P F on the y-axis. A dashed horizontal line marks zero on the P F scale. At low sample counts around 1 to 3, several points lie far below zero, reaching approximately negative 18 and negative 10. As the number of added training samples increases from about 5 onward, most points move closer to zero, fluctuating between about negative 3 and positive 3. Beyond roughly 10 samples, the points cluster tightly around zero, indicating stabilisation of P F values as more training data are added.The values of performance function for the added new samples in Example 3