Skip to main content
. Author manuscript; available in PMC: 2022 Apr 8.
Published in final edited form as: Appl Spectrosc. 2021 Aug 3;76(4):485–495. doi: 10.1177/00037028211034543

Table 2.

Out-of-sample performance for XGBoost model on five substrates.

Test sample Substrate Dye Concentration (M) Power (mW) Accuracy Precision Recall F1 AUC MCC Fraction positive
11 Moxtek R6G 1.00E-08 20 0.90 NA NA NA 0.96 NA 0.10
2 Moxtek R6G 3.00 E-03 10 0.77 0.78 0.98 0.87 0.83 0.27 0.75
3 Plasmore R6G 1.00E-08 20 0.95 0.91 0.64 0.75 0.99 0.74 0.12
4 Plasmore R6G 3.00 E-03 10 0.72 0.78 0.73 0.75 0.80 0.42 0.60
5 Plasmore R6G 1.00E-08 10 0.98 0.98 1.00 0.99 0.69 0.00 0.98
1.

For sample 1, no false positives or true negatives were generated making the precision, recall, F1, and MCC formula inapplicable.