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

Data mining of tree-based models to analyze freeway accident frequency.

Li-Yen Chang1, Wen-Chieh Chen

  • 1Graduate Institute of Transportation and Logistics, National Chia-Yi University, Taiwan. liyen@mail.ncyu.edu.tw

Journal of Safety Research
|October 29, 2005
PubMed
Summary

Classification and Regression Tree (CART) analysis identified average daily traffic and precipitation as key factors in freeway accidents. CART offers a robust alternative to traditional regression models for predicting accident frequency.

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

A Portable, Foldable Negative-Pressure Aerosol-Containment System (FNPACS) for Aerosol Control During Aerosol-Generating Procedures.

Bioengineering (Basel, Switzerland)·2026
Same author

Genomic conservation and CNS-associated intra-host variation of Nipah virus during the 1998-1999 Malaysian outbreak.

International journal of infectious diseases : IJID : official publication of the International Society for Infectious Diseases·2026
Same author

Comparative structural analysis of Nipah virus V proteins highlights V-MDA5 complex stability in pig and bat isolates.

Protein expression and purification·2026
Same author

Efficacy of Mouthwashes Delivered via Dental Unit Waterlines in Reducing Aerosolized Surrogate SARS-CoV-2.

Medical science monitor : international medical journal of experimental and clinical research·2026
Same author

Integrating Prevention and Response at the Crossroads of Henipavirus Preparedness, Hendra@30 Conference, 2024.

Emerging infectious diseases·2026
Same author

Persistent T Cell Immunity Following Nipah Virus Infection: Evidence From Malaysian Survivors.

The Journal of infectious diseases·2026

Area of Science:

  • Transportation Engineering
  • Data Mining
  • Statistical Modeling

Background:

  • Traditional statistical models like Poisson and negative binomial regression for vehicle accident analysis have limitations due to strict assumptions.
  • Violated assumptions in these models can lead to inaccurate accident likelihood estimations.
  • Classification and Regression Tree (CART) is a versatile data mining technique suitable for prediction and classification without pre-defined variable relationships.

Purpose of the Study:

  • To compare the effectiveness of CART models against traditional negative binomial regression for analyzing freeway accident frequencies.
  • To identify key determinants of vehicle accidents on national freeways using empirical data.

Main Methods:

  • Collected accident data from Taiwan's National Freeway 1 (2001-2002).

Related Experiment Videos

  • Developed and applied both Classification and Regression Tree (CART) and negative binomial regression models.
  • Analyzed the relationship between accident data and highway geometric variables, traffic characteristics, and environmental factors.

    • Main Results:

    • CART analysis revealed average daily traffic volume and precipitation as primary determinants of freeway accident frequency.
    • The study demonstrated that CART models provide accurate predictions for freeway accident occurrences.

    Conclusions:

    • Classification and Regression Tree (CART) is a valuable alternative method for analyzing freeway accident frequencies.
    • CART's ability to handle complex relationships makes it effective for traffic safety analysis.