Sign in Subscribe
Artificial intelligence

Enhancing Phishing Detection with Machine Learning: Researchers Uncover Effective Solution

Strategic Insights

— 1 min read

Researchers from the Department of Computer and Communication Engineering have made groundbreaking findings in the field of artificial intelligence, specifically in the area of phishing detection. Their study, published in the journal Applied Computational Intelligence and Soft Computing, highlighted the effectiveness of machine learning and deep learning-based frameworks in identifying phishing attempts. The research integrated diverse datasets, including phishing emails, malicious SMS messages, and URLs, to train models that demonstrated exceptional detection accuracies of up to 99.91% and 99.75%.

Key Takeaways:

  • The study investigated a machine learning and deep learning-based framework for comprehensive phishing detection across multiple phishing datasets.
  • The researchers integrated diverse datasets, including phishing emails from Kaggle, malicious SMS messages from Mendeley, and URLs from Kaggle and OpenPhish, to train the proposed models.
  • The study evaluated and compared the performance of nine state-of-the-art machine learning and deep learning models, including support vector machine (SVM) and bidirectional gated recurrent unit (BiGRU) models.
  • The experimental results demonstrated exceptional performance, with SVM and BiGRU models achieving the highest detection accuracies of 99.91% and 99.75%, respectively.
  • The study proposed an ensemble approach that synergizes the strengths of these models to enhance detection accuracy.
  • The researchers concluded that deep learning architectures generally outperform traditional machine learning techniques in identifying phishing attempts.
  • The study provided valuable insights for strengthening cybersecurity defenses in an increasingly digital world.

Statistics:

  • 99.91% detection accuracy achieved by SVM model.
  • 99.75% detection accuracy achieved by BiGRU model.
  • 9 state-of-the-art machine learning and deep learning models evaluated and compared in the study.
  • 3 diverse datasets integrated to train the proposed models: phishing emails from Kaggle, malicious SMS messages from Mendeley, and URLs from Kaggle and OpenPhish.

Sources:

  • NewsRx. Researchers from Department of Computer and Communication Engineering Describe Research in Machine Learning (Enhancing Phishing Detection: A Machine Learning Approach to Predicting Malicious Emails, URLs, and SMS Messages). Information Technology Newsweekly. November 4, 2025; p 740.
  • Enhancing Phishing Detection: A Machine Learning Approach to Predicting Malicious Emails, URLs, and SMS Messages. Applied Computational Intelligence and Soft Computing, 2025, 2025.

(The publisher for Applied Computational Intelligence and Soft Computing is Wiley. A free version of this journal article is available at https://doi-org.sdpl.idm.oclc.org/10.1155/acis/6633979.)

Read more