Research on the roof hazard factors quantification based on Chinese word segmentation and TF-IDF algorithm
- 1College of Mining, Liaoning Technical University
- 2University of Wollongong
Abstract
Roof accidents were identified as a major hazard to coal mine safety, and the analysis of 326 relevant literature items using CiteSpace reflected a growing research trend in this field. Traditional hazard identification methods were heavily dependent on expert judgment, predefined models, and manual processing, which resulted in strong subjectivity, low efficiency, and limited adaptability in handling unstructured accident texts. To address this challenge, an integrated framework combining Chinese word segmentation with the TF-IDF algorithm was proposed to automatically extract and quantify roof hazard factors. This method was applied to 115 unstructured roof accident reports to identify and quantify hazard factors, with final results indicating a 26% and 35% reduction in redundant information for direct and indirect causes, respectively. Additionally, key hazard phrases were extracted and objectively weighted by this approach. Key direct causes include mine worker, work against regulations, and untimely support measures, among others; indirect causes involve inadequate safety education and training, lack of operational regulation, and failure to implement hidden danger investigation and management.
Related
- Web-based Cellpose‑SAM segmentation enables quantification of cross-sectional area in formalin-fixed paraffin-embedded skeletal muscle sections stained with hematoxylin and eosin
- Automated segmentation and quantification of histological liver features for MASH/MASLD scoring
- A novel morphological assessment approach for abdominal aortic aneurysms: automated segmentation and measurement using a semi-supervised deep learning algorithm
- Screening glioma and glioblastoma brain tumors using dual deep learning algorithm incorporated correlative GAN and BrainNet through the probability segmentation
- Brain tumor segmentation using a combination of graph cut and optimized U-Net with genetic algorithm: An efficient hybrid framework for enhanced medical image analysis
BibTeX
@article{jia2026research,
title = {{Research on the roof hazard factors quantification based on Chinese word segmentation and TF-IDF algorithm}},
author = {Dongxu Jia and Guowei Zhang and Jun Han and Ting Ren and Chengwu Wang and Weinan Liu},
journal = {Scientific Reports},
year = {2026},
month = jul,
doi = {10.1038/s41598-026-62462-x},
url = {https://doi.org/10.1038/s41598-026-62462-x},
note = {CC BY},
}
RIS
TY - JOUR
TI - Research on the roof hazard factors quantification based on Chinese word segmentation and TF-IDF algorithm
AU - Dongxu Jia
AU - Guowei Zhang
AU - Jun Han
AU - Ting Ren
AU - Chengwu Wang
AU - Weinan Liu
PY - 2026
DA - 2026/07/01
JO - Scientific Reports
DO - 10.1038/s41598-026-62462-x
UR - https://doi.org/10.1038/s41598-026-62462-x
AB - Roof accidents were identified as a major hazard to coal mine safety, and the analysis of 326 relevant literature items using CiteSpace reflected a growing research trend in this field. Traditional hazard identification methods were heavily dependent on expert judgment, predefined models, and manual processing, which resulted in strong subjectivity, low efficiency, and limited adaptability in handling unstructured accident texts. To address this challenge, an integrated framework combining Chinese word segmentation with the TF-IDF algorithm was proposed to automatically extract and quantify roof hazard factors. This method was applied to 115 unstructured roof accident reports to identify and quantify hazard factors, with final results indicating a 26% and 35% reduction in redundant information for direct and indirect causes, respectively. Additionally, key hazard phrases were extracted and objectively weighted by this approach. Key direct causes include mine worker, work against regulations, and untimely support measures, among others; indirect causes involve inadequate safety education and training, lack of operational regulation, and failure to implement hidden danger investigation and management.
SN - 2045-2322
ER -