A Malicious Url Analysis And Detection With Effective Ocdlstm Model


  • N.Vanitha Part time Research Scholar, Department of Computer Science,Government Arts College for Men, Nandanam, Chennai, Tamilnadu
  • M. Suriakala Assistant Professor, Department of Computer science, Government Arts College for Men, Nandanam, Chennai, Tamilnadu


malicious URL, Web applications, deep learning, OCDLSTM, link structures.


Web applications are widely used across a wide range of industries thanks to their platform freedom and inexpensive operating costs. Millions of users utilise these applications on a regular basis to complete their responsibilities. Despite this, many of these applications, like phishing and fraudulent websites, are either made and maintained by hackers or are susceptible to web defacement attacks. In order to stop malware from spreading and to safeguard end users, a malware detection system is crucial. But there are a lot of solutions out there that use website content to extract features. This could harm the detection machines itself and be obscured.
Instead of analysing content, it is safer and more effective to find harmful Uniform Resource Locators (URLs). It is challenging to identify malicious software due to insufficient features and incorrect classification.A deep learning approach is recommended to find dangerous URLs across all well-known attack types. The k-best feature selection algorithm takes into account textual attributes and link structures. A model for detecting malicious URLs based on chi square dropout long short-term memory (OCDLSTM) is presented in this paper.




How to Cite

N.Vanitha, & M. Suriakala. (2023). A Malicious Url Analysis And Detection With Effective Ocdlstm Model. Chinese Journal of Computational Mechanics, (5), 264–271. Retrieved from http://jslxxb.cn/index.php/jslxxb/article/view/4358