ABSTRACT
In today’s world of enormous data, it is important to effectively extract information from the available data. This can be accomplished using feature subset selection. Feature subset selection is a method of selecting minimum number of features, with the help of which our machine can learn and accurately predict which class a particular data belongs to. In this chapter, we will introduce a new adaptive algorithm called feature selection penguin search optimization algorithm, which is a metaheuristic feature subset selection method. It is adapted from the natural hunting strategy of penguins, in which a group of penguins take jumps at random depths and come back and share the status of food availability with other penguins, and in this way, a global optimum solution is found, namely penguin search optimization algorithm. It will be combined with different classifiers to find the optimal feature subset. In order to explore the feature subset candidates, the bioinspired approach penguin search optimization algorithm generates during the process a trial feature subset that estimates its fitness value by using three different classifiers for each case: random forest, nearest neighbor, and support vector machines. However, we are planning to implement our proposed feature selection penguin search optimization algorithm on some well-known benchmark datasets collected from the UCI repository and also try to evaluate and compare its classification accuracy with some state-of-the-art algorithms.
