Natural Language Processing for Aviation Safety: Predicting Injury Levels from Incident Reports in Australia

This study investigates the application of advanced deep learning models for the classification of aviation safety incidents, focusing on four models: Simple Recurrent Neural Network (sRNN), Gated Recurrent Unit (GRU), Bidirectional Long Short-Term Memory (BLSTM), and DistilBERT. The models were eva...

Full description

Saved in:
Bibliographic Details
Main Authors: Aziida Nanyonga, Keith Joiner, Ugur Turhan, Graham Wild
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Modelling
Subjects:
Online Access:https://www.mdpi.com/2673-3951/6/2/40
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849425524897611776
author Aziida Nanyonga
Keith Joiner
Ugur Turhan
Graham Wild
author_facet Aziida Nanyonga
Keith Joiner
Ugur Turhan
Graham Wild
author_sort Aziida Nanyonga
collection DOAJ
description This study investigates the application of advanced deep learning models for the classification of aviation safety incidents, focusing on four models: Simple Recurrent Neural Network (sRNN), Gated Recurrent Unit (GRU), Bidirectional Long Short-Term Memory (BLSTM), and DistilBERT. The models were evaluated based on key performance metrics, including accuracy, precision, recall, and F1-score. DistilBERT achieved perfect performance with an accuracy of 1.00 across all metrics, while BLSTM demonstrated the highest performance among the deep learning models, with an accuracy of 0.9896, followed by GRU (0.9893) and sRNN (0.9887). Class-wise evaluations revealed that DistilBERT excelled across all injury categories, with BLSTM outperforming the other deep learning models, particularly in detecting fatal injuries, achieving a precision of 0.8684 and an F1-score of 0.7952. The study also addressed the challenges of class imbalance by applying class weighting, although the use of more sophisticated techniques, such as focal loss, is recommended for future work. This research highlights the potential of transformer-based models for aviation safety classification and provides a foundation for future research to improve model interpretability and generalizability across diverse datasets. These findings contribute to the growing body of research on applying deep learning techniques to aviation safety and underscore opportunities for further exploration.
format Article
id doaj-art-38599fbf517f4487b2b2c09e4ecabb3d
institution Kabale University
issn 2673-3951
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Modelling
spelling doaj-art-38599fbf517f4487b2b2c09e4ecabb3d2025-08-20T03:29:44ZengMDPI AGModelling2673-39512025-05-01624010.3390/modelling6020040Natural Language Processing for Aviation Safety: Predicting Injury Levels from Incident Reports in AustraliaAziida Nanyonga0Keith Joiner1Ugur Turhan2Graham Wild3School of Engineering and Technology, University of New South Wales, Canberra, ACT 2600, AustraliaCapability Systems Centre, University of New South Wales, Canberra, ACT 2610, AustraliaSchool of Science, University of New South Wales, Canberra, ACT 2612, AustraliaSchool of Science, University of New South Wales, Canberra, ACT 2612, AustraliaThis study investigates the application of advanced deep learning models for the classification of aviation safety incidents, focusing on four models: Simple Recurrent Neural Network (sRNN), Gated Recurrent Unit (GRU), Bidirectional Long Short-Term Memory (BLSTM), and DistilBERT. The models were evaluated based on key performance metrics, including accuracy, precision, recall, and F1-score. DistilBERT achieved perfect performance with an accuracy of 1.00 across all metrics, while BLSTM demonstrated the highest performance among the deep learning models, with an accuracy of 0.9896, followed by GRU (0.9893) and sRNN (0.9887). Class-wise evaluations revealed that DistilBERT excelled across all injury categories, with BLSTM outperforming the other deep learning models, particularly in detecting fatal injuries, achieving a precision of 0.8684 and an F1-score of 0.7952. The study also addressed the challenges of class imbalance by applying class weighting, although the use of more sophisticated techniques, such as focal loss, is recommended for future work. This research highlights the potential of transformer-based models for aviation safety classification and provides a foundation for future research to improve model interpretability and generalizability across diverse datasets. These findings contribute to the growing body of research on applying deep learning techniques to aviation safety and underscore opportunities for further exploration.https://www.mdpi.com/2673-3951/6/2/40natural language processingaviation safetydistilled BERTsRNN
spellingShingle Aziida Nanyonga
Keith Joiner
Ugur Turhan
Graham Wild
Natural Language Processing for Aviation Safety: Predicting Injury Levels from Incident Reports in Australia
Modelling
natural language processing
aviation safety
distilled BERT
sRNN
title Natural Language Processing for Aviation Safety: Predicting Injury Levels from Incident Reports in Australia
title_full Natural Language Processing for Aviation Safety: Predicting Injury Levels from Incident Reports in Australia
title_fullStr Natural Language Processing for Aviation Safety: Predicting Injury Levels from Incident Reports in Australia
title_full_unstemmed Natural Language Processing for Aviation Safety: Predicting Injury Levels from Incident Reports in Australia
title_short Natural Language Processing for Aviation Safety: Predicting Injury Levels from Incident Reports in Australia
title_sort natural language processing for aviation safety predicting injury levels from incident reports in australia
topic natural language processing
aviation safety
distilled BERT
sRNN
url https://www.mdpi.com/2673-3951/6/2/40
work_keys_str_mv AT aziidananyonga naturallanguageprocessingforaviationsafetypredictinginjurylevelsfromincidentreportsinaustralia
AT keithjoiner naturallanguageprocessingforaviationsafetypredictinginjurylevelsfromincidentreportsinaustralia
AT ugurturhan naturallanguageprocessingforaviationsafetypredictinginjurylevelsfromincidentreportsinaustralia
AT grahamwild naturallanguageprocessingforaviationsafetypredictinginjurylevelsfromincidentreportsinaustralia