F-score is a scale to check whether Recall and Precision are well balanced. The bigger, the better, at best 100%. It is expressed by the following formula:

• F-value = (2 x Precision x Recall) / (Precision + Recall)

For example, if “(a) Withdrawal” is set as the prediction value, and the following confusion matrix is obtained,

Recall and Precision are calculated as

• Recall = 4/(4+6)= 40%

• Precision = 4/(4+5) = 44.4%

  From this result, the F-score is

• F-value = (2 x Precision x Recall)/(Precision + Recall) = approximately 0.421

and 42.1% is displayed.