Optimization of Dataset Based on Hybrid Algorithms

Hardly a day goes by that we do not hear the emergence of a invention from here, and a scientific breakthrough from there. The world of artificial intelligence and machines embodied in technology has become an essential element in our lives, sharing most of our daily lives and helping us in most of our work and tasks, whether easy or difficult. The researchers are racing to develop programs and technologies related to machine learning and deep learning and the development of algorithms that support artificial intelligence programs, including algorithms inspired by nature developed by researchers to reduce and identify features and reduce the complexity of computational that led to obtaining high accuracy in results. Algorithms derived from natural phenomena emerged at the beginning of the nineteenth century, such as inspired algorithms (GA, PSO, ABC, GWO, BA, ALT). etc. In this article, three types of inspired algorithms (GA), (GWO) and (ANT) are used for the extraction and elimination of features that are implemented on the same dataset and making a comparison between them in terms of optimizing accuracy, runtime, features selection, and fitness value.