An Improvement in Branch and Bound Algorithm for Feature Selection
Ahood Naif Alharbi, Mohamed Yehia Dahab · 2020
Branch and bound (BB) algorithm undergoes an exponential growth in feature selection as the number of features increases, which may require, in the worst cases, exploring the whole tree looking for an optimal solution. This paper presents an enhancement in the BB algorithm for feature selection using an approximate monotonic criteria function. The enhanced version of the sub-optimal BB algorithm seeking for the solution by cutting unpromising paths and deleting multiple features at each internal tree node based on a predefined tao variable. The experiment was applied to different datasets and compared to the original BB algorithm and numerous selection methods. The results show promising results in terms of accuracy, elapsed time, and tree size.