EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

A Privacy-Preserving Hybrid Deep Neural Network for Secure Email Classification and Emotion Detection

Amit Kumar Sahu, Mukesh Kumar

International Journal of Experimental Research and Review · 30.04.2026 · 50 · 84-96

DOI: 10.52756/ijerr.2026.v50.007 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Experimental Research and Review, online erschienen am 30.04.2026. Verfasst von Amit Kumar Sahu und 1 weiteren Autor:innen. Die Arbeit nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Email communication, compared to other social media platforms, has some unique features, including long reply chains, the use of formal language, large variations in message length, the presence of irregular patterns or anomalies, and the presence of indirect relationships. Keeping these features in mind, an efficient classifier that distinguishes between spam and ham emails is proposed. To validate the performance of the proposed classifier for spam email detection, various data mining-based classification approaches are applied. The study uses the publicly available Enron email database for experimentation. To perform the experiments comprehensively, six different versions of the Enron database are integrated into the system, yielding seven database versions for experimentation. The proposed study considers various machine learning and deep learning classifiers such as Logistic Regression (LR), Convolutional Neural Networks (CNN), Random Forest (RF), Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM) and the proposed Deep Neural Network (DNN). The study uses a Bidirectional Long Short-Term Memory (BiLSTM) classifier to analyze emails and detect spam. From the experimental results, the proposed DNN–BiLSTM classifier achieves an accuracy of 96.39%, outperforming all baseline classifiers across the seven Enron dataset versions. Additionally, a hybrid AES–ECC encryption layer is incorporated for privacy-preserving email transmission.

Zitieren

APA

Sahu, A. K., & Kumar, M. (2026). A Privacy-Preserving Hybrid Deep Neural Network for Secure Email Classification and Emotion Detection. International Journal of Experimental Research and Review, 50, 84-96. https://doi.org/10.52756/ijerr.2026.v50.007

Vancouver

Sahu AK, Kumar M. A Privacy-Preserving Hybrid Deep Neural Network for Secure Email Classification and Emotion Detection. International Journal of Experimental Research and Review. 2026;50:84-96. doi:10.52756/ijerr.2026.v50.007

BibTeX
@article{Sahu2026,
  author = {Sahu, Amit Kumar and Kumar, Mukesh},
  title = {A Privacy-Preserving Hybrid Deep Neural Network for Secure Email Classification and Emotion Detection},
  journal = {International Journal of Experimental Research and Review},
  year = {2026},
  volume = {50},
  pages = {84--96},
  doi = {10.52756/ijerr.2026.v50.007}
}
Kostenlos starten →