PHISHING SITE DETECTION USING MACHINE LEARNING

COMPUTER ENGINEERING

Phishing sites represent a major concern in cybersecurity, posing significant risks to individuals and organizations alike. These deceptive websites mimic legitimate ones to trick users into divulging sensitive information such as login credentials, financial data, or personal details. The consequences of falling victim to phishing can range from identity theft and financial loss to compromising confidential information and breaching privacy. Combatting this threat requires a multi-faceted approach including user education, robust cybersecurity measures, and vigilance in detecting and reporting phishing attempts. Failure to address phishing sites effectively can result in widespread security breaches and undermine trust in online platforms and services.Numerous analysts have gone through many years making novel ways to deal with naturally distinguish phishing sites. While state of the art arrangements can convey better results, they need a ton of manual component designing and aren't great at recognizing new phishing assaults. Therefore, finding techniques that can naturally identify phishing sites and immediately oversee zero-day phishing endeavors is an open test in this field. The site page in the URL which has that contains an abundance of information that can be utilized to decide the web server's vindictiveness. AI is a powerful technique for recognizing phishing. It likewise disposes of the burdens of the past technique. The objective of this project is to train machine learning models on the dataset created to predict phishing websites. Both phishing and benign urls of websites are gathered to form a dataset and from them required url and website content-based features are extracted. The performance level of model is calculated. The workflow involves preprocessing the dataset to handle missing values and standardizing features. The Xgboost model is then trained on the labeled dataset, utilizing a gradient boosting framework to build an ensemble of decision trees. Furthermore, the system is integrated into a real-time monitoring tool that continuously assesses web pages for potential phishing threats. The tool provides timely alerts to users, enabling them to make informed decisions about the legitimacy 2 of websites they interact with. Based on the accuracy, the ability to handle missing value and speed we have used Xgboost classifier to predict the phishing website. The Xgboost based phishing site detection system demonstrates its capability to adapt to evolving threats in the dynamic landscape of online security, thereby mitigating the risks associated with phishing attacks.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.