Classifying protein-protein interaction articles using word and syntactic features

BMC Bioinformatics

Table 3 Official scores for the ACT competition.

	Run 1	Run 2	Run 3	Run 4	Run 5
TP	580	516	553	531	565
FP	417	257	376	288	398
FN	330	394	357	379	345
TN	4673	4833	4714	4802	4692
Accuracy	0.8755	0.8915	0.8778	0.8888	0.8762
Specificity	0.9181	0.9495	0.9261	0.9434	0.9218
Sensitivity	0.6374	0.5670	0.6077	0.5835	0.6209
F1 score	0.6083	0.6132	0.6014	0.6142	0.6033
MCC	0.53524	0.55306	0.52932	0.55054	0.53031
AUC iP/R	0.6591	0.6796	0.6589	0.6798	0.6537

TP, FP, FN, and TN are true positive, false positive, false negative, and true negative, respectively. MCC means Matthews’ correlation coefficient measure. AUC iP/R means the area under the interpolated precision and recall curve. F1 score and MCC evaluate the performance of binary classification. AUC iP/R evaluates system performance in terms of ranked results.

ISSN: 1471-2105