Abstract: In this fast growing modern technology driven world, the internet is one of the most important technology not only for individual users but also for organization and online business. Nowadays, there are phishers who steals sensitive information like username, password, credit card, personal data etc. Several researchers have design rule-base system for phishing detection which are credited to help people who cannot understand which Uniform Resource Locator(URL) is real or fake address. This paper concentrates on an improve Model for Detecting Phishing URL using Deep Learning. Object oriented Design Methodology was used for system architecture and structure. The support Vector Machine has been used to extract the actual and the visual link from the domain name system(DNS) and compare the actual link and the visual link if they are same. The system uses the Deep Learning model (Generated Adversarial Network) in tensor flow and Keras framework to classify Website URL dataset containing 3207 URLwebsites, 1037 are Real URL websites and 2137 are fake. The dataset was read from directory using the pandas.read_csv function. The dataset was cleaned to make sure there are no null values present. The results of the test showed accuracy of 99.8% of all input website URL classified as either Fake or Real to verify if it’s actually a Fake website or Real Website.

Keywords- Machine Learning, Deep Learning, Phishing, Generated Adversarial Network and Uniform Resource Locator.

PDF | DOI: 10.17148/IJARCCE.2021.101107

Open chat
Chat with IJARCCE