Alignment of vaccine codes using an ontology of vaccine descriptions

Table 6 Performance measures of our alignment algorithms with varying thresholds (micro-average on both reference sets)

Method	Measure	0.0	0.1	0.2	0.3	0.4	0.5	0.6	0.7	0.8	0.9	1.0
Classes	F-score	0.643	0.756	0.755	0.744	0.719	0.715	0.734	0.722	0.728	0.717	0.713
	Precision	0.592	0.745	0.745	0.746	0.747	0.756	0.923	0.939	0.967	0.970	0.968
	Recall	0.816	0.771	0.770	0.751	0.716	0.709	0.644	0.629	0.624	0.604	0.599
Metamap	F-score	0.504	0.638	0.492	0.453	0.441	0.432	0.377	0.367	0.370	0.370	0.370
	Precision	0.487	0.793	0.798	0.800	0.818	0.814	0.861	0.926	0.943	0.943	0.943
	Recall	0.828	0.590	0.426	0.381	0.367	0.357	0.285	0.279	0.277	0.277	0.277
Properties	F-score	0.847	0.932	0.932	0.932	0.920	0.920	0.879	0.875	0.870	0.857	0.857
	Precision	0.812	0.944	0.944	0.946	0.947	0.947	0.949	0.958	0.962	0.968	0.968
	Recall	0.952	0.923	0.923	0.921	0.897	0.897	0.828	0.813	0.802	0.781	0.781
Tokens	F-score	0.557	0.646	0.631	0.634	0.370	0.346	0.249	0.160	0.160	0.160	0.160
	Precision	0.538	0.788	0.845	0.895	0.877	0.888	0.952	0.950	0.933	0.933	0.933
	Recall	0.834	0.618	0.591	0.576	0.269	0.244	0.156	0.094	0.094	0.094	0.094

ISSN: 2041-1480