Changes in the Antioxidant Properties of Quince Fruit (Cydonia oblonga Miller) during Jam Production at Industrial Scale

Table 3

Adjusted parameters, performance, and relative influence (%) of Boosted Regression Trees models for antioxidant capacity (FRAP and DPPH) of quince samples.

	DPPH	FRAP
Adjusted parameters
Bag fraction	0.6	0.55
Learning rate	0.01	0.01
Tree complexity	2	2
Model performance
CV correlation	1	1
Optimal numbers of trees	31250	38000
Relative influence of polyphenols (%)
Quinic acid	16.1	14.3
Procyanidin dimer	2.5	17.2
4-O-Caffeoylquinic acid	1.4	5.5
(+) Catechin	0.7	5.0
3-p-Coumaroylquinic acid	11.4	5.5
5-O-Caffeoylquinic acid	13.6	7.7
(−) Epicatechin	2.9	0.5
5-p-Coumaroylquinic acid	25.7	28.9
Quercetin-3-O-rutinoside	0.4	4.9
Quercetin-3-O-glucoside	3.3	0.3
Kaempferol hexoside	8.6	4.6
Kaempferol rutinoside	13.5	4.1
Quercetin	0	1.6
Cumulative influence (%)	100	100