Table 2 Best subset analysis to optimally chosen the input combination of EC in ML models.

From: An improved adaptive neuro fuzzy inference system model using conjoined metaheuristic algorithms for electrical conductivity prediction

 

Combinations

R2%

R2-Ajd%

Cp

PC

1

ECt−1

0.474

0.473

90.660

0.528

2

Q/ECt−1

0.550

0.547

18.365

0.455

3

Q/ECt−4/ECt−1

0.563

0.560

6.770

0.443

4

Qt−2/Q/ECt−4/ECt−1

0.567

0.563

4.678

0.441

5

Qt−2/Q/ECt−4/ECt−2/ECt−1

0.569

0.564

4.741

0.441

6

Qt−2/Qt−1/Q/ECt−4/ECt−2/ECt−1

0.570

0.564

6.119

0.442

7

Qt−2/Qt−1/Q/ECt−4/ECt−3/ECt−2/ECt−1

0.570

0.563

8.000

0.444

  1. Significant values are in bold.