High Accuracy Phishing Detection Based on Convolutional Neural Network

Authors

  • AjithKumar Reddy K  Student, Department of Computer Science and Engineering, Nagarjuna College of Engineering and Technology, Bengaluru, Karnataka, India.
  • Darshith M P  Student, Department of Computer Science and Engineering, Nagarjuna College of Engineering and Technology, Bengaluru, Karnataka, India.
  • Divya Megha H S  Student, Department of Computer Science and Engineering, Nagarjuna College of Engineering and Technology, Bengaluru, Karnataka, India.
  • Omshree V  Student, Department of Computer Science and Engineering, Nagarjuna College of Engineering and Technology, Bengaluru, Karnataka, India.
  • Sudhakara Reddy M  Assistant Professor, Department of Computer Science and Engineering, Nagarjuna College of Engineering and Technology, Bengaluru, Karnataka, India

DOI:

https://doi.org/10.32628/IJSRST218393

Keywords:

Phishing website, anti-phishing, Logistic regression technique

Abstract

There are numerous web security dangers yet one of the significant web security issues is Phishing sites that focus on the human weaknesses instead of programming weaknesses. It tends to be depicted as the way toward acquiring the online clients to get their touchy data, for example, usernames and passwords. These days, phishing is one of the greatest regular web dangers as for the critical increase of the World Wide Web in volume over the long run. Phishing aggressors consistently utilize new (multi day) and complex procedures to beguile online clients. Thus, it is important that the counter phishing framework is ongoing and quick and furthermore influences from a shrewd phishing recognition arrangement. Here, we build up a very much established location framework which can adaptively coordinate with the changing climate and phishing sites. Our strategy is an on the web and highlight rich AI procedure to separate the phishing and real sites. Since the proposed approach removes various sorts of various highlights from URLs and pages source code, it is a totally customer side arrangement and doesn't need any assistance from the outsider. In this task, we offer a clever framework for finding phishing sites. The framework depends on an AI technique, explicitly managed learning. We have chosen the Logistic Regression strategy because of its great presentation in grouping. Our point is to acquire a better classifier by considering the attributes of phishing site and pick the better mix of them to prepare the grouped.

References

  1. Phishlabs, "2019 Phishing Trends and Intelligence Report: The Growing Social Engineering Threat" 2019, [online]availableat: https://www.phishlabs.com.
  2. K. Jain and B. B. Gupta, "An epic way to deal with ensure against phishing assaults at customer side utilizing auto-refreshed white-list," EURASIP Journal on Information Security, vol. 2016, no. 1, p. 9, 2016.
  3. Prakash, M. Kumar, R. Rao Kompella, and M. Gupta, "Phishnet: prescient boycotting to choose phishing assaults," in Proceedings of 29th IEEE Conference on Computer Communications (Infocom), pp. 1–5, Citeseer, San Diego, CA, USA, March 2010.
  4. R. M. Mohammad, L. McCluskey, and F. Thabtah, "Smart guideline based phishing sites arrangement," IET Information Security, vol. 8, no. 3, pp. 153–160, 2014.

Downloads

Published

2021-06-30

Issue

Section

Research Articles

How to Cite

[1]
AjithKumar Reddy K, Darshith M P, Divya Megha H S, Omshree V, Sudhakara Reddy M "High Accuracy Phishing Detection Based on Convolutional Neural Network" International Journal of Scientific Research in Science and Technology(IJSRST), Online ISSN : 2395-602X, Print ISSN : 2395-6011,Volume 8, Issue 3, pp.454-457, May-June-2021. Available at doi : https://doi.org/10.32628/IJSRST218393