Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: May 8, 2026

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

A metaheuristic feature selection model using bat optimization for malicious URL attack detection.

Omar Almomani1, Mahran Al-Zyoud2, Yazeed Alsarhan3

  • 1Department of Networks and Cybersecurity, Al-Ahliyya Amman University, Amman, Jordan. o.almomani@ammanu.edu.jo.

Scientific Reports
|May 6, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Security enhancement using scalable Blockchain-based Multi-Factor Authentication (BMFA).

PloS one·2026
Same author

High-fidelity machine learning models for predicting antibacterial effects of cerium oxide nanoparticles across bacterial strains.

Discover nano·2026
Same author

Advancing sustainable machining of inconel 718 through nanoparticle-enhanced coconut oil and RSM-GA optimization.

Scientific reports·2026
Same author

Deep Learning Based Computer-Aided Detection of Prostate Cancer Metastases in Bone Scintigraphy: An Experimental Analysis.

Journal of imaging·2026
Same author

Supervised machine learning intrusion detection review and multi-criteria evaluation.

Scientific reports·2026
Same author

Intelligent tool wear monitoring using XGBoost, SVR, and DNN models in NMQL environment.

Scientific reports·2026
This summary is machine-generated.

This study introduces a Bat Algorithm (BA) feature selection model to efficiently detect malicious URLs. BA significantly reduces feature dimensions, improving machine learning model accuracy for cybersecurity threat detection.

Area of Science:

  • Cybersecurity and Machine Learning
  • Bio-inspired Optimization Algorithms
  • Data Science and Feature Engineering

Background:

  • Malicious URLs (phishing, malware, spam, defacement) pose a persistent cybersecurity threat.
  • Machine learning (ML) and deep learning (DL) models struggle with high-dimensional URL features, impacting efficiency and generalization.
  • Existing security measures are often bypassed by sophisticated malicious link creation techniques.

Purpose of the Study:

  • To present a wrapper-based Bat Algorithm (BA) for effective feature selection in malicious URL detection.
  • To identify small, discriminative feature subsets to enhance ML/DL model performance.
  • To evaluate the BA model's efficacy across various ensemble ML and DL architectures.

Main Methods:

Keywords:
Bat algorithmDeep LearningFeature selectionMachine LearningURL attacks

Related Experiment Videos

Last Updated: May 8, 2026

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

  • Implemented a bio-inspired Bat Algorithm (BA) for wrapper-based feature subset selection.
  • Tested the BA model on ensemble ML (XGBoost, AdaBoost, Gradient Boosting, CatBoost, LightGBM) and DL (CNN, RNN, LSTM, CNN-LSTM) models.
  • Validated performance on the ISCX-URL-2016 and URL Phishing (2026) datasets.

    • Main Results:

    • BA achieved significant dimensionality reduction across all malicious URL types (45.91% to 67.09%).
    • Feature reduction led to consistent classification improvements on both datasets.
    • BA-enhanced LightGBM achieved top performance: 99.92% accuracy (ISCX-URL-2016) and 98.17% (URL Phishing 2026), with high ROC-AUC and efficiency.

    Conclusions:

    • The proposed BA-based feature selection is an efficient and scalable solution for malicious URL detection.
    • BA significantly enhances the accuracy and computational efficiency of ML/DL models in cybersecurity.
    • The model demonstrates strong potential for real-world implementation in cybersecurity systems.