Skip to main content
. 2021 Dec 17;11:24138. doi: 10.1038/s41598-021-03019-y

Figure 8.

Figure 8

Training time analysis of the HHO-DBN model under the ECBDL14-ROS dataset. The results portrayed that CPIO-FS with the HHO-DBN technique resulted in the least training runtime compared to the other techniques. For instance, with no FS, the HHO-DBN technique has gained a minimal runtime of 350.76 s, whereas the SVMC, LRC, and NBC methods have accomplished superior runtimes of 978.37 s, 1012.47 s, and 369.98 s, respectively. In addition, sequential CHC, the HHO-DBN approach has reached a reduced runtime of 293.87 s, whereas the SVMC, LRC, and NBC algorithms have accomplished superior runtimes of 912.40 s, 986.45 s, and 300.26 s, respectively. In addition, MR-EFS, the HHO-DBN method has reached a lesser runtime of 203.76 s, whereas the SVMC, LRC, and NBC methodologies have accomplished maximal runtimes of 864.28 s, 978.38 s, and 215.09 s, respectively. Last, CPIO-FS, the HHO-DBN manner, obtained a minimal runtime of 160.90 s, whereas the SVMC, LRC, and NBC methodologies accomplished superior runtimes of 815.89 s, 850.53 s, and 184.34 s, respectively.