Analysis of the Performance Impact of Fine-Tuned Machine Learning Model for Phishing URL Detection

Phishing leverages people’s tendency to share personal information online. attacks often begin with an email and can be used for a variety of purposes. The cybercriminal will employ social engineering techniques get the target click on link in phishing email, which take them infected website. These become more complex as hackers personalize their fraud provide convincing messages. malicious URL is advanced kind cybercrime. It might challenging even cautious users spot URLs. researchers displayed different address this challenge. Machine learning models improve detection by using URLs, web page content external features. This article presents findings experimental study that attempted enhance performance machine obtain improved accuracy two datasets are most commonly. Three distinct types tuning factors utilized, including data balancing, hyper-parameter optimization feature selection. experiment utilizes eight prevalent methods obtained from online sources, such UCI repository Mendeley repository. result demonstrates balance improves marginally, whereas hyperparameter adjustment selection significantly. algorithms combining all fine-tuned factors, outperforming existing research works. shows efficiency algorithms. For Dataset-1, Random Forest (RF) Gradient Boosting (XGB) achieve rates 97.44% 97.47%, respectively. (GB) Extreme values 98.27% 98.21%, respectively, Dataset-2.