Detecting phishing uniform resource locator(URL) by using machine learning techniques

Lim, Chian Fang (2021) Detecting phishing uniform resource locator(URL) by using machine learning techniques. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

	Text (24 Pages) Detecting phishing uniform resource locator(URL) by using machine learning techniques.pdf - Submitted Version Download (349kB)
	Text (Full Text) Detecting phishing uniform resource locator(URL) by using machine learning techniques.pdf - Submitted Version Restricted to Repository staff only Download (1MB)

Abstract

As the internet has grown in popularity, phishing websites have become more common and caused significant harm to online financial services such as online shopping and data security. Phishing is a type of fraud whereby an attacker sends a fake message or creates a phishing website to mislead web users into sharing confidential information or allowing malicious software to be installed on the victim's device. Many attackers started creating phishing websites to misled web users into thinking it’s legitimate. So, web users may be exposed to common web attacks, which might result in the loss of money, personal information, and trust from online transactions. Hence, detecting phishing websites has become a critical task that requires more examination. The most commonly used blacklist- and whitelist-based methods have shown to be ineffective. Researchers have looked into using machine learning models to detect and prevent phishing attempts. The accuracy of the prediction can be increased using machine learning methods. CatBoost based URL classifiers for detecting phishing websites are proposed in this project. The first stage is dataset will be split to 80:20 ratio to train machine learning model. The second stage involves the comparison of 3 machine learning algorithms (Logistic Regression, Random Forest, and CatBoost) the third stage involves classification of the URL's legitimacy by using CatBoost. As a result, the URL will be classified as either a phishing or a legitimate URL

Actions (login required)

View Item