A Binary Genetic Harris Hawks Optimization With Machine Learning on Detection of Phishing Url

Effective detection of the phishing URLs is possible through effective feature selection and classification technique. A potentially successful way to overcome this challenge would be nature-inspired optimization techniques using machine learning algorithm. The current investigation introduces a Binary Genetic Harris Hawks Optimization (BGHHO) method to identify phishing URLs coupled with machine learning. It builds on the standard Harris Hawks Optimization by adding a greedy crossover mechanism in an attempt to enhance exploitation. It also implements a mutation operator that is based on the concept of differential evolution to improve exploration and population diversity. BGHHO is used to make feature selection and the features that have been selected are classified with a K-Nearest Neighbors (KNN) classifier. The phishing URL dataset is used to evaluate the performance of the approach with performance metrics like average fitness value, feature selection rate, and classification accuracy.