Page 305 - Data Science Algorithms in a Week

286 Jose M. Prieto

antioxidant capacities of essential oils of known chemical composition in both assays
using an artificial neuronal network (ANN) with an average error of only 1-3% (Cortes-
Cabrera & Prieto, 2010). Later, MusaAbdullah, & Al-Haiqi (2015) successfully modeled
the popular DPPH assay with ANNs but using a camera as imaging system instead the
conventional colorimetry reader.

Table 1. Prediction of Antioxidant properties of Natural Products using ANNs

Natural Product Input Output Reference
Apple pomace, Apple pomace, orange or Peroxide values of (Ozturk et al.,
orange and potato potato peels content. sunflower oil samples 2012)
peels
Ascorbic acid in Thermal treatments Ascorbic acid content (Zheng, et al.,
green Asparagus parameters 2011)
Bananas Total phenols (Guiné et al.,
2015.)
Bayberry juice Red, green, and blue (RGB) Anthocyanins, ascorbic (Zheng et al.,
intensity values acid, Total phenols, 2011.)
flavonoids, and
antioxidant activity
Centella asiatica Selected shifts in the 1H DPPH radical (Maulidiani et al.,
Nuclear Magnetic Resonance scavenging activity. 2013)
spectra corresponding to 3,5-
O-dicaffeoyl-4-O-
malonilquinic acid (irbic
acid), 3,5-di-O-caffeoylquinic
acid, 4,5-di-O-caffeoylquinic
acid, 5-O-caffeoylquinic acid
(chlorogenic acid), quercetin
and kaempferol.
Cinnamon, clove, Colorimetry of the reaction % Scavenged DPPH (Musa et al., 2015)
mung bean, red
bean, red rice,
brown rice, black
rice and tea extract
Clove bud essential Peroxide concentration; Autooxidation of (Misharina et al.,
oil, Ginger, pimento thiobarbituric acid reactive polyunsaturated fatty 2015)
and black pepper substances; diene conjugate acids in linseed oil.
extracts content; content of volatile
compounds formed as
products of unsaturated fatty
acid peroxide degradation;
and composition of methyl
esters of fatty acids.
Commercial teas Total flavonoids, total total antioxidant activity (Cimpoiu et al.,
catechines and total 2011)
methylxanthines