International Journal of Advanced Research in Computer and Communication Engineering

A monthly peer-reviewed online and print journal

ISSN Online 2278-1021
ISSN Print 2319-5940

Since 2012

Abstract: Now days everything is digitalized and it’s difficult to secure our data. Web security is a major issue because everything is connected through internet. The common way of launching an internet attack is by using malicious URLs. Hackers or Intruders leaks billions of confidential data every year by using malicious websites The traditional way of detecting these kinds of malicious URLs or websites is by the use of a web database called Blacklists. There are many URL shortening methods and Domain Generation Algorithms are available which make it difficult to detect the newly generated malicious URLs. To increase the efficiency and avoid database dependency we proposed the Machine learning approach to detect the malicious URLs. In machine learning approach there are so many algorithms available for the classification and feature extraction from that we select the best method that is Random forest which will gave us more accurate result than the past ones. So our proposed system will increase the efficiency and gave accurate prediction whether the URL entered is malicious or benign by using a well-defined dataset. It is also possible to implement the machine learning application in a proxy server or any network traffic controller system.

Keywords: Malicious URL; Random forest; Machine learning; Blacklist; Detection; prediction; URL; benign


PDF | DOI: 10.17148/IJARCCE.2019.8247

Open chat
Chat with IJARCCE