In this paper, we present a hybrid evolutionary approach used for feature extraction based on using an optimized version of the crow search algorithm with an optimized spider monkey algorithm–hybrid crow–spider monkey algorithm. Crows are scientifically regarded as one of the most intelligent animals in nature. This nature of theirs for food foraging is replicated in a crow search algorithm, and using it for feature selection from datasets is explained in the optimized crow search algorithm. By observing the intelligent foraging behaviour of spider monkeys, a fitness-controlled optimization algorithm has been defined that picks the best set of features just as a swarm of spider monkeys would pick the most probable locations for getting food. This algorithm takes the dataset as input and returns a modified dataset with all instances of selected features. Through this hybrid process, we can perform our computations using fewer features as an important set of features is segregated by the optimized crow search algorithm. Further, the selected set is filtered by an optimized spider monkey algorithm to give a very highly optimized set of features which being less in number provide almost similar or greater than the original classification accuracy and take lesser computation time, thus making us able to reduce and eliminate computational overheads as shown in the results demonstrated towards the end of this paper.