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 Videos

Support vector machines for spam categorization.

H Drucker1, D Wu, V N Vapnik

  • 1AT&T Labs-Research, Red Bank, NJ 07701, USA.

IEEE Transactions on Neural Networks
|February 7, 2008
PubMed
Summary
This summary is machine-generated.

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

Primary structure and functional characterization of a soluble, alternatively spliced form of B7-1.

Journal of immunology (Baltimore, Md. : 1950)·2000
Same author

Regional analysis of divorce in China since 1980.

Demography·2000
Same author

Role of Raf-1 conserved region 2 in regulation of Ras-dependent Raf-1 activation.

Biochemical and biophysical research communications·2000
Same author

[Loss of heterozygosity and microsatellite instability in laryngeal carcinoma near p16 gene].

Zhonghua yi xue yi chuan xue za zhi = Zhonghua yixue yichuanxue zazhi = Chinese journal of medical genetics·2000
Same author

Acute effects of growth hormone in alcohol-fed rats.

Alcohol and alcoholism (Oxford, Oxfordshire)·2000
Same author

Impaired protein synthesis induced by acute alcohol intoxication is associated with changes in eIF4E in muscle and eIF2B in liver.

Alcoholism, clinical and experimental research·2000
Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013
Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011
Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011
Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011
Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011
Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011
See all related articles

Support vector machines (SVMs) and boosting decision trees offer effective spam email classification. SVMs demonstrated superior performance with binary features and significantly reduced training times compared to other algorithms.

Area of Science:

  • Machine Learning
  • Data Mining
  • Computational Intelligence

Background:

  • Email spam classification is a critical task in cybersecurity.
  • Various machine learning algorithms are employed for this purpose.
  • Comparing algorithm performance across different feature set sizes is essential.

Purpose of the Study:

  • To evaluate the efficacy of Support Vector Machines (SVMs) for spam email classification.
  • To compare SVMs against Ripper, Rocchio, and boosting decision trees.
  • To analyze algorithm performance on datasets with varying dimensionality.

Main Methods:

  • Utilized four classification algorithms: Support Vector Machines (SVMs), Ripper, Rocchio, and boosting decision trees.
  • Tested algorithms on two datasets: one with 1000 features and another with over 7000 features.

Related Experiment Videos

  • Evaluated performance based on accuracy, speed, and training time, particularly with binary features.

    • Main Results:

    • Support Vector Machines (SVMs) showed the best performance when utilizing binary features.
    • Both boosting trees and SVMs achieved acceptable test accuracy and speed across both datasets.
    • SVMs exhibited significantly shorter training times compared to other tested algorithms.

    Conclusions:

    • Support Vector Machines (SVMs) are a highly efficient choice for spam email classification, especially with binary features.
    • SVMs offer a favorable balance of accuracy, speed, and notably reduced training time.
    • The study highlights SVMs as a competitive and time-efficient algorithm for large-scale email filtering.