Table 2

Best performance achieved by each algorithm for classifying clinical handover incidents

Algorithm	Accuracy (%)		Precision		Recall		F-measure		Area under curve		Feature selection methods		Training set size
Algorithm	R	E	R	E	R	E	R	E	R	E	R	E	R	E
Naïve Bayes	86.29	91.53	0.84	0.87	0.90	0.98	0.87	0.92	0.93	0.97	1, 2	1, 2	600	300
Support Vector Machine (linear)	84.27	89.11	0.85	0.86 0.87	0.84	0.94 0.92	0.84	0.90 0.89	0.90	0.93 0.92	1, 2	1, 2 1, 2	300	300 500
Support Vector Machine (radial-basis function)	83.87	88.71	0.81	0.82	0.90	0.99	0.85	0.90	0.92	0.97	1, 2	1, 2	600	600

E, performance tested against tested against expert-classified handover incidents; R, performance tested against reporter-classified incidents.