A Fast Logdet Divergence Based Metric Learning Algorithm for Large Data Sets Classification

Table 4

Performance comparison with the state-of-the-art online metric learning methods on the large data sets.

Data set	LEGO		Proposed algorithm
	Precision	Time (s)	Compressed dimension	Precision	Time (s)
Letter-Recognition	0.9348	1.2316	10	0.9351	6.3149
Spambase	0.7829	0.7659	20	0.9033	0.9543
Isolet	0.9156	166.9784	20	0.9167	3.2437
Semeion	0.9190	2.4171	20	0.9210	0.1829
Mfeat	0.9375	43.9377	20	0.9496	0.4436