Animal-borne acoustic data alone can provide high accuracy classification of activity budgets

Table 2 Information extracted from 61 reviewed classifications (54 articles, including our study)

Devices	N studies	Species			Devices weight (g)		Sampling frequency (Hz)		Number of activities		Algorithm								Global accuracy (%)		Additional information
Devices	N studies	Terr	Aqua	Fly	Range	N	Range	N	Range	N	RF	DT	SVM	DA	NN	KNN	MM	Mix	Range	N	Additional information
Accelerometer alone	34	25	8	1	[2.0–371]	16	[0.007–100]	33	[2–12]	34	9	5	5	4	3	2	1	4	[73.3–100.0]	34	Energy expenditure
Accelerometer + other(s)	16	10	4	2	[15.5–160]	7	[0.02–100]	16	[2–8]	16	4	5	2	2	1	1	0	1	[71.8–99.0]	15	Energy expenditure + potentially others
GPS	7	4	0	3	[4.6–5500]	4	[0.003–1]	7	[2, 3]	7	3	2	0	1	0	0	1	0	[65.0–94.4]	7	Geographical position
Acoustic recorder	3	1	0	2	[16.2–41]	2	[16,000–22,050]	3	[3–19]	3	1	1	0	0	0	1	0	0	[94.1–98.5]	2	Biophony, geophony, anthropophony
Pressure sensor (depth)	1	0	1	0	–	0	0.011	1	3	1	0	0	0	0	0	0	1	0	94.0	1	Diving profile

Other devices deployed concomitantly to accelerometers: GPS (n = 4), gyroscope (n = 3), magnetometer (n = 1), gyroscope + magnetometer (n = 5), pressure (n = 1), gyroscope + pressure (n = 1), magnetometer + acoustic (n = 1). Species categorisation: terr. terrestrial, aqua. aquatic, fly. flying. Algorithms: RF random forest, DT decision tree, SVM support vector machine, DA discriminant analysis, NN neural network, KNN K-nearest neighbour, MM Markov Model, mix combination of several algorithms [26, 32, 52,53,54, 68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115]

ISSN: 2050-3385