Study and Analysis of Machine Learning Models for Detection of Phishing URLs
Shreyas Sanjeev Desai, Sahil Salunkhe, Rashmi J. Deshmukh, Sheetal S. Zalte · 2023
The increase in use of the Internet has led to the increase in usage of many e-commerce applications, social media, and a lot of other websites where it is required by the user to give his/her sensitive credentials while using. Along with the websites, there is also an increase in “Smart” devices in the market which have access to a lot of your physical belongings in your day-to-day life as well. Phishing is a problem which has been in the Internet world for a long time. Stealing a user's sensitive information without his consent by deceiving him about the genuineness of the website is called phishing. Most times, the users are not aware that they are visiting a website which is trying to steal information, and unknowingly they give away a lot of personal information. This leads to Internet fraud and a lot of cybercrimes. To avoid this, we have proposed a phishing detection method with a minimal number of features and higher accuracy as possible. We have used a number of machine learning models for this purpose and compared their performance based on some performance metrics. We have taken phishing URLs from the PhishTank dataset and a list of legitimate URLs from the Alexa dataset.