A Learning-Based Approach for Biomedical Word Sense Disambiguation

Table 7

The averaged evaluation results from Wang et al. [9].

	Micro-avg.			Macro-avg.
	Precision	Recall	F1	Precision	Recall	F1
RULE-MAJORITY	72.2	62.39	66.94	27.77	46.67	29.32
RULE-SP	74.09	64.03	68.69	29.77	53.81	32.2
RULE-SPSENT	72.94	63.03	67.63	30.22	54.76	32.93
C&C	73.82	63.79	68.44	30.51	53.59	33.43
ENJU	72.98	63.06	67.66	31.35	55	34.61
ENJU-Genia	73	63.08	67.68	30.11	53.42	32.97
Minipar	73.02	63.1	67.69	30.19	53.56	33.1
Stanford	73.67	63.66	68.3	31.17	56.35	34.35
Stanford-Genia	73.48	63.5	68.13	30.61	55.61	33.78
ML	82.69	82.69	82.69	27.01	27.84	27.37
RELATION	75.24	63.99	69.16	31.97	55.61	34.8
HYBRID	83.8	83.8	83.8	57.56	49.72	49.9