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

Finding the right decision tree's induction strategy for a hard real world problem.

M Zorman1, V Podgorelec, P Kokol

  • 1Laboratory for System Design, Faculty of Electrical Engineering and Computer Science, University of Maribor, Smetanova 17, SI-2000 Maribor, Slovenia. milan.zorman@uni-mb.si

International Journal of Medical Informatics
|August 24, 2001
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

Walking is the form of physical activity people with osteoarthritis in the hip or knee choose and maintain after two years.

Osteoarthritis and cartilage open·2025
Same author

Breathing Easy in Maricopa County, Arizona: Empowering K-12 Schools With a Novel Nurse Coaching Model to Facilitate Implementation of a County-Wide School-Based Stock Albuterol Inhaler Program.

The Journal of school nursing : the official publication of the National Association of School Nurses·2025
Same author

One leg testing in hip and knee osteoarthritis: A comparison with a two-leg oriented functional outcome measure and self-reported functional measures.

Osteoarthritis and cartilage·2024
Same author

Retroperitoneal Fat Necrosis in Response to an Episode of Acute Pancreatitis.

Case reports in gastrointestinal medicine·2021
Same author

Reply by Authors.

The Journal of urology·2020
Same author

The Burden of Functional Disabilities for Middle-Aged and Older Adults in the United States.

The journal of nutrition, health & aging·2019

For complex medical data, traditional decision tree methods struggle. An evolutionary approach offers the best balance of accuracy, sensitivity, and tree size for building effective decision trees in real-world scenarios.

Area of Science:

  • Medical Informatics
  • Machine Learning
  • Computational Statistics

Background:

  • Decision trees are valuable tools in medicine but face limitations with complex, real-world datasets.
  • Traditional induction methods may not yield optimal results for challenging medical problems.

Purpose of the Study:

  • To evaluate various strategies for building univariate decision trees.
  • To identify the most effective induction strategy for complex medical data, specifically orthopaedic fracture cases.

Main Methods:

  • Compared four classical decision tree induction approaches.
  • Implemented a hybrid approach combining neural networks and decision trees.
  • Utilized an evolutionary approach for decision tree construction.
  • Tested methods on a dataset of 2637 orthopaedic fracture cases with 23 attributes.

Related Experiment Videos

Main Results:

  • All tested approaches encountered challenges with accuracy, sensitivity, or decision tree size.
  • The evolutionary approach demonstrated superior performance compared to classical and hybrid methods.
  • The evolutionary approach provided the best compromise for decision tree building on this difficult dataset.

Conclusions:

  • Traditional decision tree induction methods are insufficient for certain complex medical problems.
  • The evolutionary approach represents a promising strategy for developing accurate and efficient decision trees in medical informatics.
  • Further research into evolutionary algorithms for medical decision support systems is warranted.