The Use of Machine Learning Approaches for the Diagnosis of Acute Appendicitis

Table 4

Results of gradient boosting tree analysis.

Accuracy: 95.31%	True 1	True 0	Total	Class precision (%)
Pred. 1	55 (TP); Correct Decision	2 (FP); Type I error	57 (P′)	96.49
Pred. 0	4 (FN); Type II error	67 (TN); Correct Decision	71 (N′)	94.36
Total	59 (P)	69 (N)	128 (P + N)
Class recall	93.22%	97.10%

TP: true positives, TN: true negatives, FN: false negatives, and FP: false positives. Precision: it is the ratio of correctly predicted positive samples to the number of samples estimated in the positive class. Recall: it is the ratio of correctly predicted positive samples the ratio to the number of samples in the true positive class.