A Noise-Resilient Attention Guided Contrastive Learning Framework for Phishing Email Detection

doi:10.23940/ijpe.26.06.p1.297308

Abstract

Abstract: Phishing emails are a serious cybersecurity risk because they take advantage of people's weaknesses to obtain private information. Both conventional heuristics and classical machine learning techniques are undermined by their intrinsically noisy, obfuscated, and constantly changing character. Because they are unable to learn invariant, discriminative representations and deliberately suppress noise, even the most advanced deep learning models have limited robustness. In this paper, we provide a novel noise-augmented, attention-guided contrastive learning framework for reliable phishing email detection in order to overcome these difficulties. However, we introduce two crucial novelties: (i) the noise augmentation policy, producing perturbed views of an email to account for variability and obfuscation in the wild, and (ii) an attention-guided contrastive learning method to focus on more informative features and discard noisy ones in contrastive representation learning. By leveraging this synergy, the model can obtain class-discriminative and noise-invariant embeddings, guaranteeing accurate identification even in the face of extreme adversarial perturbations. Comprehensive evaluations on the publicly accessible Phishing Email Detection dataset show that the proposed framework routinely achieves accuracy gains of at least 1.5% and 1.22%, respectively, over twelve cutting-edge baselines. These findings confirm the efficacy and superiority of the suggested method for acquiring reliable, noise-resistant representations for precise phishing detection in practical settings.

Key words: phishing email detection, noise-resilient learning, contrastive learning, attention mechanism, cybersecurity, deep learning

El Yazid Gueddoudj, Abdelouahab Attia, and Azeddine Chikh. A Noise-Resilient Attention Guided Contrastive Learning Framework for Phishing Email Detection [J]. Int J Performability Eng, 2026, 22(6): 297-308.

Add to citation manager EndNote|Reference Manager|ProCite|BibTeX|RefWorks

References

[1] Uddin M.A., Mahiuddin M., andSarker I.H., 2026. An explainable transformer-based model for phishing email detection: A large language model approach.Computer Networks, 112061.
[2] Ahmed R., andSumesh E.P., 2026. An intelligent phishing email detection system using ensemble methods and explainable AI.Knowledge-Based Systems, 115388.
[3] Bountakas P., andXenakis C., 2023. Helphed: hybrid ensemble learning phishing email detection.Journal of Network and Computer Applications, 210, 103545.
[4] Bergholz A., De Beer J., Glahn S., Moens M.F., Paaß G., andStrobel S., 2010. New filtering approaches for phishing email. Journal of Computer Security,18(1), pp. 7-35.
[5] Fang Y., Zhang C., Huang C., Liu L., andYang Y., 2019. Phishing email detection using improved RCNN model with multilevel vectors and attention mechanism.IEEE Access, 7, pp. 56329-56340.
[6] Atawneh S., andAljehani H., 2023. Phishing email detection model using deep learning.Electronics, 12(20), 4261.
[7] Khalid A., Hanif M., Hameed A., Ashraf Z., Alnfiai M.M., andAlnefaie S.M.M., 2024. Logitriblend: A novel hybrid stacking approach for enhanced phishing email detection using ml models and vectorization approach.IEEE Access, 12, pp. 193807-193821.
[8] Daniel M.A., Chong S.C., Chong L.Y., andWee K.K., 2025. Optimising phishing detection: A comparative analysis of machine learning methods with feature selection. Journal of Informatics and Web Engineering,4(1), pp. 200-212.
[9] Do N.Q., Selamat A., Krejcar O., andFujita H., 2025. Detection of malicious urls using temporal convolutional network and multi-head self-attention mechanism.Applied Soft Computing, 169, 112540.
[10] Sarker I.H.,2023. Machine learning for intelligent data analysis and automation in cybersecurity: current and future prospects. Annals of Data Science,10(6), pp. 1473-1498.
[11] Le-Khac P.H., Healy G., andSmeaton A.F., 2020. Contrastive representation learning: A framework and review.IEEE Access, 8, pp. 193907-193934.
[12] Niu Z., Zhong G., andYu H., 2021. A review on the attention mechanism of deep learning.Neurocomputing, 452, pp. 48-62.
[13] Butt U.A., Amin R., Aldabbas H., Mohan S., Alouffi B., andAhmadian A., 2023. Cloud-based email phishing attack using machine and deep learning algorithm. Complex & Intelligent Systems,9(3), pp. 3043-3070.
[14] Fatima R., Fareed M.M.S., Ullah S., Ahmad G., andMahmood S., 2024. An optimized approach for detection and classification of spam email’s using ensemble methods. Wireless Personal Communications,139(1), pp. 347-373.
[15] Zamir A., Khan H.U., Iqbal T., Yousaf N., Aslam F., Anjum A., andHamdani M., 2020. Phishing web site detection using diverse machine learning algorithms. the Electronic Library,38(1), pp. 65-80.
[16] Li Q., Cheng M., Wang J., andSun B., 2020. LSTM based phishing detection for big email data. IEEE Transactions on Big Data,8(1), pp. 278-288.
[17] Rashed S., andOzcan C., 2026. A novel dual-layer deep learning architecture for phishing and spam email detection.Electronics, 15(3), 630.
[18] Kaddoura S., Alfandi O., andDahmani N., 2020. A spam email detection mechanism for english language text emails using deep learning approach. In2020 IEEE 29th International Conference on Enabling Technologies: Infrastructure for Collaborative Enterprises (WETICE), pp. 193-198.
[19] Safran M., andMusleh A., 2025. PhishingGNN: phishing email detection using graph attention networks and transformer-based feature extraction.IEEE Access, 13, pp. 131390-131399.
[20] Vrahatis A.G., Lazaros K., andKotsiantis S., 2024. Graph attention networks: a comprehensive review of methods and applications.Future Internet, 16(9), 318.
[21] Sanjalawe Y., Fraihat S., Kurdi G., andAllehyani B., 2025. A cognitive BDI-BiLSTM hybrid framework for interpretable email threat detection.IEEE Access, 13, pp. 208192-208211.
[22] Champa A.I., Rabbi M.F., Eishita F.Z., andZibran M.F., 2025. Catch me if you see: using visual cue and explanatory feedback to enhance human phishing detection.IEEE Access, 13, pp. 193249-193272.
[23] Heiding F., Schneier B., Vishwanath A., Bernstein J., andPark P.S., 2024. Devising and detecting phishing emails using large language models.IEEE Access, 12, pp. 42131-42146.
[24] Yi X., Li L., Zhang J., andJia Z., 2026. Heterogeneous federated learning for imbalanced phishing email detection.Cybersecurity, 9(1), 10.
[25] Hu H., Wang X., Zhang Y., Chen Q., andGuan Q., 2024. A comprehensive survey on contrastive learning.Neurocomputing, 610, 128645.
[26] Robinson J., Chuang C.Y., Sra S., andJegelka S., 2020. Contrastive learning with hard negative samples.Arxiv Preprint Arxiv:2010.04592.
[27] Wang Z., Liu L., Weston S.R.F., Tian S., andLi P., 2024. On learning discriminative features from synthesized data for self-supervised fine-grained visual recognition. InEuropean Conference on Computer Vision, pp. 101-117.
[28] Jing L., Vincent P., LeCun Y., andTian Y., 2021. Understanding dimensional collapse in contrastive self-supervised learning.Arxiv Preprint Arxiv:2110.09348.
[29] Caron M., Misra I., Mairal J., Goyal P., Bojanowski P., andJoulin A., 2020. Unsupervised learning of visual features by contrasting cluster assignments.Advances in Neural Information Processing Systems, 33, pp. 9912-9924.
[30] Chen T., Kornblith S., Norouzi M., andHinton G., 2020. A simple framework for contrastive learning of visual representations. InInternational Conference on Machine Learning, pp. 1597-1607.
[31] Gulli A., andPal S., 2017. Deep Learning with Keras. Packt Publishing Ltd.
[32] Abadi M., Barham P., Chen J., Chen Z., Davis A., Dean J., Devin M., Ghemawat S., Irving G., Isard M., andKudlur M., 2016. {TensorFlow}: a system for {large-scale} machine learning. In12th USENIX Symposium on Operating Systems Design and Implementation (OSDI 16), pp. 265-283.
[33] Kaggle, Phishing Email Detection, https://www.kaggle.com/datasets/subhajournal/phishingemails, accessed on March 1, 2026.
[34] Xu S., Li Y., andWang Z., 2017. Bayesian multinomial naïve bayes classifier to text classification. InInternational Conference on Multimedia and Ubiquitous Engineering, pp. 347-352.
[35] Meyer, D. and Wien, F.T., 2001. Support vector machines. R news,1(3), pp.23-26.
[36] Hosmer Jr D.W., Lemeshow S., andSturdivant R.X., 2013. Applied Logistic Regression. John Wiley & Sons.
[37] Alzubaidi L., Zhang J., Humaidi A.J., Al-Dujaili A., Duan Y., Al-Shamma O., Santamaría J., Fadhel M.A., Al-Amidie M., andFarhan L., 2021. Review of deep learning: concepts, CNN architectures, challenges, applications, future directions.Journal of Big Data, 8(1), 53.
[38] Muzaffar S., andAfshari A., 2019. Short-term load forecasts using LSTM networks.Energy Procedia, 158, pp. 2922-2927.
[39] Liu G., andGuo J., 2019. Bidirectional LSTM with attention mechanism and convolutional layer for text classification.Neurocomputing, 337, pp. 325-338.
[40] Acheampong F.A., Nunoo-Mensah H., andChen W., 2021. Transformer models for text-based emotion detection: a review of Bert-based approaches. Artificial Intelligence Review,54(8), pp. 5789-5829.