
Applications of Data Mining in Organizational Behavior ◾ 249
e output of testing neural network is depicted in Figure12.6. At this stage,
Mean Square Error (MSE), Mean Absolute Error (MAE), . . . indicators are com-
puted by software and the real values are compared with the predicted values. As it
is observed, these two values are relatively similar, which in turn implies the accu-
racy of the model in prediction.
Finally, sensitivity analysis is used for investigating the inuence of input fac-
tors on the satisfaction computed by neural network. As illustrated in Figure12.7,
employees’ satisfaction has the highest and lowest sensitivity to ...