Persian Named Entity Recognition by Gray Wolf Optimization Algorithm

Table 2

Comparing three gold-standards (GS 1, GS 2, and GS 3) with each other.

Gold-standard		GS1		GS2		GS3		GS-average
Compared with		GS2	GS3	GS1	GS3	GS1	GS2
Partial	Precision	0.584	0.450	0.735	0.541	0.630	0.633	0.595
	Recall	0.703	0.605	0.595	0.621	0.442	0.544	0.585
	F1	0.638	0.516	0.658	0.578	0.519	0.585	0.582
Exact	Precision	0.481	0.320	0.670	0.418	0.533	0.530	0.492
	Recall	0.579	0.430	0.543	0.480	0.373	0.455	0.477
	F1	0.525	0.367	0.6	0.447	0.439	0.489	0.478

The bold values show the highest and lowest results.