A Hybrid Fully Homomorphic Encryption Based Privacy Preserving DeepCNN Framework for Email Spam Classification
Keywords:
E-mail, Spam Filtering, Fully Homomorphic Encryption, Deep Convolution Neural Network, FHE based Privacy Preserving ProtocolAbstract
The E-mail filtering systems have been designed mainly to identify spam and either block it or put it in the spam folder. An effective adaboost sequential classification based ensemble method used where the differential grading weighting schemes are used to classify the results more accurately as nonspam emails are delivered to inbox and spam emails are stored in filtering system folder. The execution of the spam action would be impaired by the lack of effective strategies to deal with threats to the security of the spam filter. The need to apply deep learning to spam filtering to leverage its many layers of processing and multiple levels of abstraction to learn data representations. In this Paper, the effect of implementing the filtering spam emails in email spam filtering system entirely overcome by addressed fully homomorphic encryption based privacy preserving and deep convolution neural network. In deep convolution neural network, an adapted transnet is proposed to increase the spam classification accuracy and reduce communication complexities on the transit layer. The proposed fully homomorphic encryption based protocol concealed with privacy preserving mechanism. Initially, firewall check the spam emails to find IP address, media access control address and domain name. An unauthenticated spam mails are stored in deny list and delivered to trash. In fully homomorphic encryption based privacy preserving and deep convolution neural network, an authenticated spam emails are encrypted using fully homomorphic encryption algorithm then encrypted emails are fed into deep learning classifier. The deep learning classification method classify the encrypted emails as spam and nonspam. The encrypted spam and nonspam emails are send to mail server decrypted and deliver emails to spam folder and inbox. The experiments results prove that the proposed method is better than existing email text spam classification method
Downloads
Metrics
References
Vijayasekaran, G., & Rosi, S. (2018), “Spam and Email Detection in Big Data Platform using Navies Bayesian Classifier”, International Journal of Computer Sci. and Mobile Computing, Vol. 7, No. 4, pp. 53-58, ISSN: 2320-088X.
Gauri Jain., Manisha., & Basant Agarwal, (2016), “An Overview of RNN and CNN Techniques for Spam Detection in Social Media”, International Journal of Advanced Research in Computer Science and Software Engineering, Vol. 6, No.10, pp.126-138, ISSN: 2277 128X.
Ajay Shrestha., & Ausif Mahmood, (2019), “Review of Deep Learning Algorithms and Architectures”, IEEE Access, Vol. 7, pp. 53040-53065, ISSN: 2169-3536.
Muhammd Imran Tariq., Nisar Ahmed Memon., & Muhammad Imran, (2020), “A Review of DeepLearning Security and Privacy Defensive System”, Mobile Information System, PP 1-18 ISSN: 1574-017X.
Zvika Brakerski., Craig Gentry., & Vinod Vaikuntanathan, (2014), “Fully Homomorphic Encryption without Bootstrapping”, ACM Transactions on Computation Theory, Vol. 6, No.3, pp. 1-36.
Benavides E., Sanchez S., & Fuertes, W. (2020), “Classification of Phishing Attack Solutions by employing DeepLearning techniques : A Systematic literature Review”, Springer, pp. 51-64.
Ajinkya Gulunjkar., & Rakesh Rathi (2020), “A Machine Learning Approach Based Spam Filtering With the Use of Neural Network”, ITEE Journal, Vol.9, No.4, ISSN: 2306-708X.
Carlos Laorden., Xabier Ugarte Pedrero., & Igor Santos, (2014), “Study on the Effectiveness of anomaly detection for spam filtering”, Information Sciences, Elsevier, Vol.277, pp. 421-444, ISSN: 0020-0255.
Khedr, A., Gulak, G., & Vaikuntanathan, V. (2014), “SHIELD: Scalable Homomorphic Implementations of Encrypted Data Classifiers”, IEEE Transactions on Computers, Vol.65, No.9, ISSN: 2848-2858.
Ahmad Al Badawi., Louie Hoang., & Chan Fook Mun (2020), “PrivFT: Private and Fast Text Classification with Homomorphic Encryption”, IEEE Access, Vol.8, pp. 22654-22656.
Renukadevi, R. et al. “An Improved Collaborative User Product Recommendation System Using Computational Intelligence with Association Rules.” Communications on Applied Nonlinear Analysis (2024): n. pag. https://doi.org/10.52783/cana.v31.1243
“TransNet: Minimally Supervised Deep Transfer Learning for Dynamic Adaptation of Wearable Systems”, ACM Transactions on Design Automation, Vol.26, No.1, pp 1-31, ISSN: 1084-4309.
Ammara Zamir, Hikmat Ullah Khan & Waqar Mehmood, “A feature-centric spam email detection model using diverse supervised machine learning algorithms”,pp 633-657,
Erkin, Z., Veugen, T., & Toft, (2012), “Generating private recommendations efficiently using homomorphic encryption and data packing,” IEEE Transactions on Information Forensics and Security, Vol. 7, No. 3, pp. 1053–1066.
Badsha, S., X. Yi., & Khalil, I. (2016) , “A practical privacy-preserving recommender system,” Data Science and Engineering, vol. 1, no. 3, pp.161–177.
Aono, Y., Hayashi, T., & Wang, L. (2017), “Privacy-preserving deep learning via additively homomorphic encryption”, IEEE Trans. Inf. Forensics Secur., Vol. 13, 1333-1345.
Al-Rubaie, M., Chang. J.M. (2019), “Privacy-preserving machine learning: Threats and solutions”, IEEE Secur. Priv., Vol. 17, 49-58.
Mohamad .M & Selamat .A, “An evaluation on the efficiency of hybrid feature selection in spam email classification”, IEEE Communications and Control Technology (I4CT), pp. 227-231, 2015.
Karthika Renuka .D, Visalakshiand .P, Rajamohana .SP, “An Ensembled Classifier for Email Spam Classification in Hadoop Environment”, Appl. Math. Inf. Sci., Vol. 4, No. 11, pp. 1123-1128, 2017.
Izzat Alsmadi & Ikdam Alhami, “Clustering and Classification of email contents”, Journal of King Saud and Information Sciences, pp. 46-57, 2015.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
You are free to:
- Share — copy and redistribute the material in any medium or format
- Adapt — remix, transform, and build upon the material for any purpose, even commercially.
Terms:
- Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
- No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
