Skip to main navigation Skip to search Skip to main content

Optimized Deep Learning Based Phishing Email Detection Using BERT and Hill Climbing Algorithm

  • Akshat Gaurav
  • , Brij B. Gupta
  • , Arcangelo Castiglione
  • , Shavi Bansal
  • , Kwok Tai Chui

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Prevention of cybersecurity risks in contemporary communication systems depends on phishing email detection. This work presents an optimal deep learning method using the Hill Climbing (HC) algorithm for hyperparameter optimization and BERT for feature extraction to improve phishing detection. Using a Kaggle dataset, the model was trained with balanced precision, recall, and F1-scores for phishing and safe emails with a 95% accuracy. In terms of decreased loss and improved generalization, a comparative study encompassing GRU, LSTM, RNN, Logistic Regression, and SVM showed the suggested method’s excellence. The results highlight how well feature extraction combined with optimization methods could help to identify phishing emails in practical environments.

Original languageEnglish
Title of host publicationComputational Data and Social Networks - 13th International Conference, CSoNet 2024, Proceedings
EditorsJanos Kertesz, Bo Li, Thepchai Supnithi, Akkharawoot Takhom
Pages258-269
Number of pages12
DOIs
Publication statusPublished - 2025
Event13th International Conference on Computational Data and Social Networks, CSoNet 2024 - Bangkok, Thailand
Duration: 16 Dec 202418 Dec 2024

Publication series

NameLecture Notes in Computer Science
Volume15417 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference13th International Conference on Computational Data and Social Networks, CSoNet 2024
Country/TerritoryThailand
CityBangkok
Period16/12/2418/12/24

Keywords

  • BERT
  • Hill Climbing Algorithm
  • Phishing Detection

Fingerprint

Dive into the research topics of 'Optimized Deep Learning Based Phishing Email Detection Using BERT and Hill Climbing Algorithm'. Together they form a unique fingerprint.

Cite this