MALICIOUS URL DETECTION SYSTEM USING COMBINED SVM AND LOGISTIC REGRESSION MODEL

In this paper, we have discussed our skills in training and testing of a malicious uniform resource locator detecting system. there have been many trends in the technology and security which inspire our analysis. First, being online has grown to be a progressively dangerous place for many people. In 2011, according to Symantec, there is an increase in internet attacks by 36%. This interprets into roughly 4,500 attacks on a daily basis. Second, there is a vital growth in personal and enterprise use of mobile for the internet. In 2012, according to the State of quality Survey by Symantec, it was noted that after principally proscribed by IT, mobile phones are getting used currently by much immeasurable staff throughout the planet. Online vulnerabilities are on the rise with the employment of smart phones and devices for every personal and skilled usage. This research targets machine learning answers to identify malicious uniform resource locators employing a combination of URL lexical options, payload size, and python supply options. we tend to use Support Vector Machine with apolynomial kernel and logistic regression to attain an accuracy of 98%.