Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

Root-Locus Method

Root-Locus Method

A cruise control system in a car is designed to maintain a specified speed automatically by adjusting the gas pedal. The system continuously measures the vehicle's speed and makes fine adjustments to the pedal to achieve this goal. The root locus method is particularly useful for understanding how the cruise control system's behavior changes under varying conditions, such as when the car goes uphill, downhill, or faces strong wind resistance.
This system can be represented by a block diagram,...

You might also read

Related Articles

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

Sort by

Same author

Pathology and Genetics in a Global Cohort of Parkinsonian Disorders.

JAMA neurology·2026

Same author

Pathology and genetics in a global cohort of Parkinsonian Disorders.

medRxiv : the preprint server for health sciences·2026

Same author

Astrocyte Proximity Protects Synapses From Human Amyloid-Beta Induced Degeneration in a Mouse Ex Vivo Model of Early Alzheimer's Disease.

The European journal of neuroscience·2026

Same author

Mapping the Invisible: How Cerebral Vascular Phenotypes Shape Cognitive Aging.

Stroke·2026

Same author

Antidepressants and bleeding risk: Expert consensus from the Association of Medicine and Psychiatry.

Journal of psychosomatic research·2026

Same author

Nanoscopic tau aggregates are not shared intermediates but disease-specific entities across tauopathies.

Cell reports·2026

Same journal

Study of adaptation parameters of merging zones for freeway interchange ramps considering CAV & AV mixed traffic environments.

Traffic injury prevention·2026

Same journal

Behavioral determinants of motorcyclist personal protective equipment (PPE) use in five European cities.

Traffic injury prevention·2026

Same journal

Impact of grade designs of long mountainous freeway tunnel with crest vertical curve on traffic safety at tunnel portals.

Traffic injury prevention·2026

Same journal

Associations between blood ethanol concentration, clinical impairment test results, and traffic accident involvement among apprehended drivers.

Traffic injury prevention·2026

Same journal

Crash severity determinants in Saudi Arabia's Eastern Province: Evidence from logistic regression and neural networks.

Traffic injury prevention·2026

Same journal

Seeing the automation malfunction before it happens: designing information cues that support driver situation awareness.

Traffic injury prevention·2026

See all related articles

Search research articles

Related Experiment Video

Updated: Jul 3, 2026

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

Finding and understanding pedal misapplication crashes using a deep learning natural language model.

Max Bareiss¹, Colin Smith¹, Hampton C Gabler¹

¹Department of Biomedical Engineering, Virginia Tech, Blacksburg, Virginia.

Traffic Injury Prevention

|December 7, 2021

Summary

This summary is machine-generated.

A BERT model accurately identifies pedal misapplication (PM) crashes from narratives, achieving 95.7% accuracy and processing data 353 times faster than manual review.

Keywords:

BERT NLP NMVCCS Pedal misapplication deep learning

More Related Videos

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

Related Experiment Videos

Last Updated: Jul 3, 2026

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

Area of Science:

Natural Language Processing
Machine Learning in Transportation Safety
Automotive Accident Analysis

Background:

Pedal misapplication (PM) is a critical factor in vehicle crashes.
Manual review of crash narratives is time-consuming and resource-intensive.
Existing methods for identifying PM from text data are limited.

Purpose of the Study:

To develop and validate a BERT-based system for automated identification of PM crashes.
To assess the accuracy and efficiency of the automated system compared to manual review.
To enable large-scale analysis of crash data using natural language processing.

Main Methods:

Utilized a BERT language model trained on crash narratives from NMVCCS and North Carolina databases.
Employed a keyword search algorithm to select relevant cases for training and testing.
Classified narratives into categories: no PM, PM by vehicle 1, 2, or 3.
Validated the model's performance using AUC ROC and comparison with manual review.

Main Results:

The BERT model achieved a strong AUC ROC of 0.9835, indicating excellent generalization.
The system correctly identified PM in 95.7% of crash narratives.
The correct vehicle involved in PM was identified in 95.9% of cases.
Automated processing was 353 times faster than manual researcher review.

Conclusions:

Automated interpretation of crash narratives using deep learning is feasible and highly accurate.
This approach significantly reduces research time and enables analysis of large, previously inaccessible datasets.
The technique can be extended to extract other critical vehicle, occupant, or environmental data from crash narratives.