Based on URL Feature Extraction Identify Malicious Website Using Machine Learning Techniques

A phishing attack is the simplest way to obtain sensitive information from users. The aim of the phishers is to acquire critical information like username, password, bank account details and other personal information. With the development of Internet technology, network security is under different threats. Especially attackers can spread malicious uniform resource locators (URLs) to carry out attacks such as phishing and spam. The research on malicious URL detection is significant for defending against this attack. Some existing detection methods are easy to cover by attackers. We design a malicious URL detection model based on Machine Learning Techniques to solve these problems. Cyber security persons are now looking for reliable and stable detection techniques for phishing websites detection. This propose system deals with machine learning technology for the detection of phishing URLs by extracting and analysing various feature of legitimate and phishing URLs. Decision Trees, random forest and support vector machine algorithms are used to detect phishing websites or unsecure websites. The aim of the paper is to detect phishing URLs as well as cut down to the best machine learning algorithm by comparing the accuracy rate, false positive and false negative rate of each algorithm. This paper analyses the structural feature of the URL of the Phishing websites extracts 12 kinds of features and uses four machine learning algorithms for training and use the best-performing algorithm as our model to identify unknown URLs.