Phishing Threats: How Deep Learning Can Keep You Safe

By: Jampula Navaneeth1

1Vel Tech University, Chennai, India

2International Center for AI and Cyber Security Research and Innovations, Asia University, Taiwan Email: navaneethjampula@gmail.com

Abstract

As internet is growing the online fraud activities are being increasing simultaneously. In that phishing is one of the main cybercrime activities. We all are habituated of using mobile phones, basically phishing attacks can be started from getting close to the persons and sharing them fake links. Now a days we are seeing there are many detection models are implemented for phishing attacks. So in this article we are going to see what are the threats that we can face in phishing attacks and how deep learning will help in protecting from phishing attacks.

Introduction

Phishing is the most utilized attack in social engineering. In this way the phisher seeks to gather identity information of the user with an intent to use the information maliciously against the user or its organization. The easiest thing to do with these attackers is to fall prey after opening a given Web site with a fake links. This article’s objective is to review the current state of machine learning algorithms in identification of social engineering attacks linked to phishing using Deep Learning approaches [1].

Phishing Attacks

Phishing is an attack whereby the attacker tries to deceitfully acquire information, such as usernames, passwords and credit card details from a syncretistic identity in electronic communication. The unsuspecting public is lured by means of communication purported to be from favourite social websites, auction sites, online payments process or IT administrator. Phishing emails can contain links redirecting viewers to sites which may harm malware [2].

Purpose of Deep Learning

Deep learning is a subset of machine learning which is designed to implicate human mind. In phishing detection deep learning help in analysing the large amount of data. As opposed to the conventional knowledge representation-based systems, deep learning models are capable of processing large volumes of information and recognize signs of phishing [3]. Because of this characteristic of adaptive learning, deep learning proves beneficial in the rapidly evolving context of cyber threats.

Phishing Detection using Deep Learning

DL architecture is created based on the neural networks, with some capabilities to identify the hidden information in the analysed data through the gradual build-up of levels. For the detection of phishing, the usage of DL approach has been increasing in recent years as DL technologies advanced [3]. While the amount of required data and the training time are larger for DL compared to the traditional ML method, this method is capable of determining the features from raw data signals without prior knowledge. Recently, different types of methods that are based on DL have been applied to improve the performance of classification for the detection of phishing [4]. This includes various steps to detect phishing as shown in the below figure 1.

Figure 1: Working process of phishing detection

Advantage of Deep Learning in Phishing Detection

  • High Accuracy: There is a high success rate of accuracy in deep learning models used in the detection of phishing attempts accompanied by few false positives and few false negatives.
  • Adaptability: The feature of learning from new data makes the deep learning models versatile and capable of handling new techniques in phishing, making it provide better defence to the new methods.
  • Automated Analysis: It is possible to employ deep learning taking into account the fact that this approach allows for the identification of phishing attacks in large data in real time [5], [6].

Challenges

By using deep learning, we have improved many aspects of this process; but with that comes its challenges. Creating and sustaining models for deep learning is a laborious task that requires strong amounts of computational power, in addition to being well-informed [7]. Furthermore, a large labelled dataset is required (ideally of good quality) for training the models well. Privacy issues associated with the data employed to train these models must also be a priority for organizations [1], [4].

Conclusion

Phishing threats still remain as a major security concern to organizations and individuals in an increasingly connected world, but deep learning is shown as a way to improve the security mechanisms. As a result of using the capability of deep learning, organizations can identify and stop the occurrence of phishing attacks effectively and efficiently. Since the cybercriminals’ threat is becoming more diverse, the continuous learning capabilities of the deep learning models will be very helpful in protecting the individuals and organizations from phishing threats.
Thus, the enhancement of deep learning utilization in the cybersecurity action plans will be critical in future to counter the increase in cyber risks and guarantee safe and secure digital spaces.

References

  1. E. Benavides, W. Fuertes, S. Sanchez, and M. Sanchez, “Classification of Phishing Attack Solutions by Employing Deep Learning Techniques: A Systematic Literature Review,” in Developments and Advances in Defense and Security, Á. Rocha and R. P. Pereira, Eds., Singapore: Springer, 2020, pp. 51–64. doi: 10.1007/978-981-13-9155-2_5.
  2. V. Bhavsar, A. Kadlak, and S. Sharma, “Study on Phishing Attacks,” International Journal of Computer Applications, vol. 182, pp. 27–29, Dec. 2018, doi: 10.5120/ijca2018918286.
  3. O. K. Sahingoz, E. BUBEr, and E. Kugu, “DEPHIDES: Deep Learning Based Phishing Detection System,” IEEE Access, vol. 12, pp. 8052–8070, 2024, doi: 10.1109/ACCESS.2024.3352629.
  4. N. Q. Do, A. Selamat, O. Krejcar, E. Herrera-Viedma, and H. Fujita, “Deep Learning for Phishing Detection: Taxonomy, Current Challenges and Future Directions,” IEEE Access, vol. 10, pp. 36429–36463, 2022, doi: 10.1109/ACCESS.2022.3151903.
  5. S. Maurya and A. Jain, “Deep learning to combat phishing,” Journal of Statistics and Management Systems, vol. 23, no. 6, pp. 945–957, Aug. 2020, doi: 10.1080/09720510.2020.1799496.
  6. P. Pappachan, N. S. Adi, G. Firmansyah, and M. Rahaman, “Deep Learning-Based Forensics and Anti-Forensics,” in Digital Forensics and Cyber Crime Investigation, CRC Press, 2024.
  7. M. Rahaman, B. Chappu, N. Anwar, and P. K. Hadi, “Analysis of Attacks on Private Cloud Computing Services that Implicate Denial of Services (DoS),” vol. 4, 2022.
  8. Aldweesh, A., Alauthman, M., Al Khaldy, M., Ishtaiwi, A., Al-Qerem, A., Almoman, A., & Gupta, B. B. (2023). The meta-fusion: A cloud-integrated study on blockchain technology enabling secure and efficient virtual worlds. International Journal of Cloud Applications and Computing (IJCAC), 13(1), 1-24.
  9. M. Casillo, F. Colace, B. B. Gupta, A. Lorusso, F. Marongiu and D. Santaniello, “Blockchain and NFT: a novel approach to support BIM and Architectural Design,” 2022 International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies (3ICT), Sakheer, Bahrain, 2022, pp. 616-620, doi: 10.1109/3ICT56508.2022.9990815.
  10. P. Chaudhary, B. B. Gupta, K. T. Chui and S. Yamaguchi, “Shielding Smart Home IoT Devices against Adverse Effects of XSS using AI model,” 2021 IEEE International Conference on Consumer Electronics (ICCE), Las Vegas, NV, USA, 2021, pp. 1-5, doi: 10.1109/ICCE50685.2021.9427591.

Cite As

Navaneeth J. (2024) Phishing Threats: How Deep Learning Can Keep You Safe, Insights2Techinfo, pp.1

78740cookie-checkPhishing Threats: How Deep Learning Can Keep You Safe
Share this:

Leave a Reply

Your email address will not be published.