Feature Selection for Classification on High- Dimensional Data Using Swarm Optimization Algorithms

Feature selection plays a crucial role in classification by identifying relevant features. Bio-inspired algorithms, such as genetic algorithms, particle swarm optimization, and ant colony optimization, have gained popularity due to their ability to mimic natural processes and explore high-dimensional feature spaces efficiently. The analysis includes evaluating classification accuracy, and computational efficiency as performance metrics. The experimental setup involves dataset selection, choosing a classification algorithm, implementing bio-inspired feature selection, evaluation, parameter tuning, and comparative analysis. The findings provide valuable insights into the effectiveness and efficiency of bio-inspired algorithms for feature selection in classification. The project focuses on reducing high- dimensional datasets to low-dimensional ones using feature selection through classification with swarm optimization algorithms. It aims to validate the effectiveness of the Jellyfish Search algorithm (JSA), Horse Herd Optimization algorithm (HOA), and Binary Bat algorithm (BBA) on real-world datasets. The objectives include evaluating the algorithms' performance, conducting a comprehensive performance analysis, execution of classifiers like RF, DT, KNN, Voting Classifier and SVM, obtaining and tuning the results, and publishing the findings. The research outcomes contribute to the knowledge of feature selection for classification and provide guidance in selecting appropriate algorithms for specific classification tasks.