Analysis of Machine Learning Techniques for Sentinel-2A Satellite Images

Table 8

Group 1 area and percentages of land changes for machine learning in Sentinel-2A satellite (2016).

No.	Name	RF		KNN		DTC		SVM		Mahalanobis		Normal Bayes
		Area, km^2	%	Area, km^2	%	Area, km^2	%	Area, km^2	%	Area, km^2	%	Area, km^2	%
1	High land	49807900	2.65	59806000	3.18	47879400	2.55	35915200	3.80	29958400	2.27	41708400	2.22
2	Land area	725187700	38.58	766080400	40.76	701479200	37.32	818676800	86.70	527685000	39.91	750954500	39.95
3	Agricultural area	322988700	17.18	326778200	17.38	337729300	17.97	396037600	41.94	334353400	25.29	270312300	14.38
4	Built-up area	552998800	29.42	495807400	26.38	551666100	29.35	420439800	44.53	838959700	63.46	407204500	21.66
5	Mountains	214799500	11.43	198756400	10.57	226929200	12.07	171002900	18.11	109322700	8.27	363201000	19.32
6	Bare land	13905900	0.74	32460300	1.73	14005500	0.75	37616300	3.98	39409500	2.98	46308000	2.46
Total		1879688500	100.00	1879688700	100.00	1879688700	100.00	944277500	100.00	1322045300	100.00	1879688700	100.00