Predicting moose behaviors from tri-axial accelerometer data using a supervised classification algorithm

Table 2 Samples sizes for each individual and behavior

Animal ID	Foraging	Lying_o	Lying_u	Ruminating	Running	Standing	Walking	Total
Stella	12,308	3763	19,247	13,242	18	5408	1661	55,647
Babe	15,732	2791	15,832	9787	89	8462	2590	55,283
Wilma	6970	1383	20,242	15,274	16	9465	1474	54,824
Sky	13,126	1354	15,733	15,999	49	5226	1715	53,202
Shiner	7397	1390	23,916	11,784	9	6359	2176	53,031
Cayenne	13,149	732	12,568	12,927	7	8536	2197	50,116
Roxanne	9392	1386	16,271	10,006	41	7243	3716	48,055
Minnie	6349	1678	22,927	8434	28	5253	1988	46,657
Winnie	4326	99	3426	4078	0	1128	266	13,323
Vicky	3205	49	4034	4451	37	709	236	12,721
Mattis	753	0	2729	2036	50	3897	727	10,192
Idun	865	0	2614	929	13	1894	637	6952
Lily	2832	0	1277	2203	7	362	203	6884
Olivia	2303	10	1392	1930	0	1035	121	6791
Total	98,707	14,635	162,208	113,080	364	64,977	19,707	473,678

Number of labeled 3-s accelerometer data intervals for each behavior and individual moose used to train the random forest model classifying animal-borne accelerometer data into seven discrete behaviors

ISSN: 2050-3385