Applying Randomness Effectively Based on Random Forests for Classification Task of Datasets of Insufficient Information

Procedure 1

Procedure:

Begin

 Check if the grid search could be effective by generating trial random forests;

 /*  |𝐴|: the number of conditional attributes */

 R  := the double of the first integer less than log₂|𝐴| + 1;

 I = 100; F = 1000; D = 25;

 Do

   For  t = I  to  F  by  increasing   D

    /* Generate random forests of t trees in which R attributes are picked

    randomly to split each node */

    Generate Random_forests (R, t);

   End  For;

   R  := the first integer larger than R/2;

  Until   R = 1;

End.