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

Autonomous evolutionary algorithm in medical data analysis.

Matej Sprogar1, Miha Sprogar, Matjaz Colnaric

  • 1Faculty of Electrical Engineering and Computer Science, University of Maribor, Slovenia. matej.sprogar@uni-mb.si

Computer Methods and Programs in Biomedicine
|March 8, 2006
PubMed
Summary

This study introduces an autonomous evolutionary algorithm for building decision trees with minimal human input. It self-adapts to medical datasets, offering general solutions or indicating when data is too complex for analysis.

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

Evolution in medical decision making.

Journal of medical systemsยท2002
See all related articles

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Bioinformatics

Background:

  • Decision tree construction is crucial for medical data analysis.
  • Existing algorithms often require significant human expertise and parameter tuning.
  • Overfitting remains a challenge in medical machine learning applications.

Purpose of the Study:

  • To present an autonomous evolutionary algorithm for decision tree construction.
  • To evaluate the algorithm's performance and self-adaptation capabilities on medical datasets.
  • To develop a method for predicting dataset difficulty or impossibility of analysis.

Main Methods:

  • Utilizes a co-evolving environment with non-standard implicit fitness evaluation.
  • Incorporates self-adaptation of evolutionary parameters for autonomous behavior.

Related Experiment Videos

  • Applies the algorithm to diverse medical datasets to assess its generalizability.

    • Main Results:

    • The autonomous algorithm demonstrates minimal need for human interaction.
    • It exhibits self-monitoring and self-adjustment capabilities.
    • Successfully produces generalizable solutions or abstains from solutions for ill-posed problems, indicating dataset difficulty.

    Conclusions:

    • The presented autonomous evolutionary algorithm offers a novel approach to decision tree construction in medicine.
    • Its self-adaptive nature allows it to handle complex and potentially problematic datasets effectively.
    • The algorithm's ability to predict analysis difficulty aids in resource allocation and study design.