Table 3.

Supervised machine learning approaches in plant electrochemical impedance spectroscopy.

Description	Type	Algorithm	Metric	Performance	Size	Year	Ref
Apple mouldy core	Classification	SVM	Accuracy	0.94	98	2016	(Yu et al., 2016)
Avocado ripeness	Classification	SVM	Accuracy	0.90	100	2018	(Islam et al., 2018)
Grapefruit freeze damage	Classification	MLP	Accuracy	1.00	180	2022	(Romero Fogué et al., 2022)
Lemon freeze damage	Classification	MLP	Accuracy	1.00	10	2019	(Ochandio Fernández et al., 2019)
Oil palm basal stem rot	Classification	LDA	Accuracy	0.86	240	2022	(Khaled et al., 2022)
Olive variety	Classification	MLP	Accuracy	1.00	90	2020	(Luna et al., 2020)
Orange freeze damage	Classification	MLP	Accuracy	1.00	270	2018	(Serrano-Pallicer et al., 2018)
Plant tissue discrimination	Classification	MLP	Accuracy	1.00	100	2020	(Cavalieri and Bertemes-Filho, 2020)
Rice seed vigor	Classification	LDA	Accuracy	0.90	100	2021	(Feng et al., 2021)
Strawberry ripeness	Classification	MLP	F1	0.72	923	2021	(Ibba et al., 2021)
Strawberry ripeness	Classification	MLR	Accuracy	0.773	150	2017	(González-Araiza et al., 2017)
Tangerine freeze damage	Classification	MLP	Accuracy	1.00	270	2021	(Aparisi et al., 2021)
Tomato ripeness	Classification	LDA	Accuracy	0.88	240	2019	(Li et al., 2019)
Wood chips	Classification	KNN	Accuracy	0.91	NA	2020	(Tiitta et al., 2020)
Apple moisture content	Regression	PLS	R 2	0.88	140	2018	(Reyes et al., 2018)
Apple soluble solids content	Regression	ELM	R 2	0.908	160	2015	(Guo et al., 2015)
Banana soluble solids content	Regression	LR	R 2	0.716	90	2014	(Jamaludin et al., 2014)
Crop leaf nitrogen content	Regression	MLR	R 2	0.94	111	2020	(Basak et al., 2020b)
Date acidity	Regression	MLP	R 2	0.938	800	2022	(Mohammed et al., 2022)
Durian dry matter content	Regression	PLS	RMSE	4.63%	120	2013	[Kuson and Terdwongworakul, 2013]
Korla pear hardness	Regression	NFS	R 2	0.911	61	2022	(Yu et al., 2022)
Korla pear soluble solids content	Regression	GRNN	R 2	0.974	300	2020	(Lan et al., 2020)
Leaf moisture content	Regression	MLR	R 2	0.959	28	2021	(Hao et al., 2021)
Lettuce Chlorophyll content	Regression	MLR	RMSE	1.05 μg/L	70	2021	(Chowdhury et al., 2021)
Lime moisture content	Regression	PLS	R 2	0.934	82	2016	(Huong and Teerachaichayut, 2016)
Melon sugar content	Regression	ELM	R 2	0.887	480	2021	(Liu et al., 2021b)
Palm fruitlet oil content	Regression	LR	RMSE	5.71%	90	2022	(Chin-Hashim et al., 2022)
Peach firmness	Regression	CART	RMSE	1.59 N	200	2022	(Ivanovski et al., 2022)
Peach firmness	Regression	LR	MSE	0.67	200	2020	(Ivanovski et al., 2020)
Persimmon soluble solids content	Regression	LS-SVM	RMSE	0.97°Brix	105	2017	(Liu and Guo, 2017)
Pineapple sugars content	Regression	MLP	R 2	0.973	54	2016	(Conesa et al., 2016)
Sea buckthorn soluble solids	Regression	MLR	R 2	0.648	NA	2022	(Li et al., 2022)
Sweet potato moisture content	Regression	PLS	R 2	0.44	80	2018	(Reyes et al., 2018)
Tomato leaf nitrogen content	Regression	MLR	R 2	0.8374	35	2017	(Meiqing et al., 2017)
Tomato leaf phosphor content	Regression	MLR	R 2	0.864	34	2016	(Meiqing et al., 2016)
Tomato leaf potassium content	Regression	MLR	R 2	0.8561	34	2016	(Jinyang et al., 2016)

The encountered algorithms are $K$ -nearest neighbors (KNN), (least-squares-) Support Vector Machine ((LS-)SVM), Linear Discriminant Analysis (LDA), Classification And Regression Trees (CART), Extreme Learning Machines (ELM), Neuro-Fuzzy System (NFS), Generalized regression neural network (GRNN), Partial Least Squares (PLS), Multi-layer Perceptron (MLP), and (Multivariate) Linear Regression (M)LR. When multiple classification problems or algorithms were considered, a single one was selected and reported per reference. If multiple algorithms were used, only the highest-performing one was reported. Data prepossessing steps are not reported.