Skip to main content
. 2018 Jun 15;8:9161. doi: 10.1038/s41598-018-27337-w

Table 5.

Results of the LSTM with TI model trained with different combinations of feature categories.

Models Accuracy PPIA SPIA
Full Model 0.9906 ± 0.0043 0.9894 ± 0.0074 0.9912 ± 0.0039
Model without CDR 0.9557 ± 0.0121 0.9470 ± 0.0273 0.9565 ± 0.0125
Model without GDS 0.9900 ± 0.0057 0.9832 ± 0.0103 0.9906 ± 0.0054
Model without FAQ 0.9615 ± 0.0140 0.9429 ± 0.0262 0.9624 ± 0.0143
Model without CDR, GDS 0.9699 ± 0.0094 0.9637 ± 0.0154 0.9708 ± 0.0091
Model without CDR, FAQ 0.7191 ± 0.0282 0.6867 ± 0.0471 0.7207 ± 0.0278
Model without GDS, FAQ 0.9689 ± 0.0144 0.9620 ± 0.0229 0.9698 ± 0.0141
Model without CDR, GDS, FAQ 0.7148 ± 0.0355 0.6868 ± 0.0437 0.7147 ± 0.0353

Note that the results presented here are mean values and the standard deviation values of the metrics for the 10-fold cross validation, and the performances of each fold are provided in the Supplementary Tables S8–S14.