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 Concept Videos

Classification of Systems-I01:26

Classification of Systems-I

Linearity is a system property characterized by a direct input-output relationship, combining homogeneity and additivity.
Homogeneity dictates that if an input x(t) is multiplied by a constant c, the output y(t) is multiplied by the same constant. Mathematically, this is expressed as:
Classification of Systems-II01:31

Classification of Systems-II

Continuous-time systems have continuous input and output signals, with time measured continuously. These systems are generally defined by differential or algebraic equations. For instance, in an RC circuit, the relationship between input and output voltage is expressed through a differential equation derived from Ohm's law and the capacitor relation,
Classification of Signals01:30

Classification of Signals

In signal processing, signals are classified based on various characteristics: continuous-time versus discrete-time, periodic versus aperiodic, analog versus digital, and causal versus noncausal. Each category highlights distinct properties crucial for understanding and manipulating signals.
A continuous-time signal holds a value at every instant in time, representing information seamlessly. In contrast, a discrete-time signal holds values only at specific moments, often denoted as x(n), where...

You might also read

Related Articles

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

Sort by
Same author

Gated recurrent unit model for forecasting greenhouse gas concentrations with uncertainty quantification.

Frontiers in artificial intelligence·2026
Same author

A labelled dataset of healthy and diseased common bean (<i>Phaseolus vulgaris</i>) from Tanzania.

Data in brief·2026
Same author

Enhanced SQL injection detection using chi-square feature selection and machine learning classifiers.

Frontiers in big data·2025
Same author

Enhancing detection of common bean diseases using Fast Gradient Sign Method-trained Vision Transformers.

Frontiers in artificial intelligence·2025
Same author

Irish potato imagery dataset for detection of early and late blight diseases.

Data in brief·2025
Same author

Banana Leaves Imagery Dataset.

Scientific data·2025

Related Experiment Videos

Adaptive class-aware feature selection for high-dimensional and imbalanced multi-class network intrusion detection.

Joseph P Mchina1, Neema Mduma1, Ramadhani S Sinde1

  • 1Computational and Communication Science and Engineering (CoCSE), The Nelson Mandela African Institution of Science and Technology (NM-AIST), Arusha, Tanzania.

Frontiers in Big Data
|July 8, 2026
PubMed
Summary

This study introduces an Adaptive Class-Aware Feature Selection (ACAFS) framework to improve machine learning-based network intrusion detection systems (ML-NIDS). ACAFS enhances detection of rare attacks by adaptively selecting features, significantly reducing dimensionality while boosting performance.

Keywords:
adaptive feature selectionmachine learningminority attack detectionmutual informationnetwork intrusion detection systems (NIDS)

Related Experiment Videos

Area of Science:

  • Cybersecurity
  • Machine Learning
  • Network Security

Background:

  • High-dimensional data and class imbalance challenge Machine Learning-based Network Intrusion Detection Systems (ML-NIDS).
  • Existing feature selection methods often overlook minority attack classes due to reliance on global relevance and fixed feature counts.
  • This leads to reduced sensitivity towards critical, rare attack categories in network traffic.

Purpose of the Study:

  • To propose the Adaptive Class-Aware Feature Selection (ACAFS) framework for multi-class intrusion detection.
  • To address limitations of conventional feature selection in ML-NIDS, particularly concerning class imbalance and minority attack detection.
  • To develop a data-driven approach for adaptive feature selection that enhances detection of rare but critical network threats.

Main Methods:

  • Developed the Adaptive Class-Aware Feature Selection (ACAFS) framework incorporating a data-driven adaptive feature count mechanism using permutation null hypothesis testing.
  • Implemented a Class-Aware Composite Mutual Information scoring strategy to preserve minority-class discriminative information.
  • Employed a coordinated two-stage feature selection process combining statistical filtering with XGBoost-based model refinement.

Main Results:

  • ACAFS significantly reduced feature dimensionality by 70.3% on the CSE-CIC-IDS2018 dataset (from 74 to 22 features).
  • Achieved high detection performance with a Two-Stage CNN: 99.39% accuracy, 99.40% F1-score, and 0.09% FPR.
  • Demonstrated superior minority attack detection, reaching 98.59% recall for Web_Attacks on CSE-CIC-IDS2018 despite severe class imbalance.

Conclusions:

  • The ACAFS framework effectively reduces feature redundancy and improves the detection of minority attack classes in ML-NIDS.
  • Adaptive and class-aware feature selection enhances intrusion detection performance across diverse network environments, as validated on CSE-CIC-IDS2018 and SUNE datasets.
  • ACAFS offers a robust solution for building more sensitive and accurate network intrusion detection systems, even with imbalanced data.