Table 3 Performance of different algorithms on real-world data sets.
From: Adaptive soft sensor using stacking approximate kernel based BLS for batch processes
Name of data set | BLS | KBLS | AKBLS | |||
|---|---|---|---|---|---|---|
\({\text{R}}^{{2}}\) | Time | \({\text{R}}^{{2}}\) | Time | \({\text{R}}^{{2}}\) | Time | |
Penicillin dataset 1 | 0.99774 | 0.08096 | 0.99938 | 0.07596 | 0.99924 | 0.02797 |
Penicillin dataset 2 | 0.99538 | 0.16213 | 0.99823 | 0.19988 | 0.99782 | 0.09594 |
Penicillin dataset 3 | 0.99132 | 0.96596 | 0.99742 | 2.55625 | 0.99918 | 1.50113 |
Penicillin dataset 4 | 0.98983 | 5.21843 | 0.99355 | 16.2907 | 0.9934 | 9.97005 |
Steel industry energy consumption | 0.98718 | 5.60765 | 0.99788 | 15.3896 | 0.99793 | 7.38505 |
Flight price prediction | 0.97042 | 9.66289 | 0.98484 | 27.125 | 0.98512 | 14.0166 |
Electric power consumption | 0.91691 | 8.2057 | 0.9288 | 19.2953 | 0.9357 | 11.5036 |
Remaining useful lifetime prediction | 0.99645 | 16.7048 | 0.9994 | 104.052 | 0.99944 | 27.8417 |
