Characterizing accident narratives with word embeddings: Improving accuracy, richness, and generalizability

A Review of Data Mining Strategies by Data Type, with a Focus on Construction Processes and Health and Safety Management

Increasingly, information technology facilitates the storage and management of data useful for risk analysis and event prediction. Studies on data extraction related to occupational health and safety are increasingly available; however, due to its variability, the construction sector warrants special attention. This review is conducted under the research programs of the National Institute for Occupational Accident Insurance (Inail).

OBJECTIVES

The research question focuses on identifying which data mining (DM) methods, among supervised, unsupervised, and others, are most appropriate for certain investigation objectives, types, and sources of data, as defined by the authors.

METHODS

Scopus and ProQuest were the main sources from which we extracted studies in the field of construction, published between 2014 and 2023. The eligibility criteria applied in the selection of studies were based on the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA). For exploratory purposes, we applied hierarchical clustering, while for in-depth analysis, we used principal component analysis (PCA) and meta-analysis.

RESULTS

The search strategy based on the PRISMA eligibility criteria provided us with 63 out of 2234 potential articles, 206 observations, 89 methodologies, 4 survey purposes, 3 data sources, 7 data types, and 3 resource types. Cluster analysis and PCA organized the information included in the paper dataset into two dimensions and labels: "supervised methods, institutional dataset, and predictive and classificatory purposes" (correlation 0.97-8.18 × 10-1; p-value 7.67 × 10-55-1.28 × 10-22) and the second, Dim2 "not-supervised methods; project, simulation, literature, text data; monitoring, decision-making processes; machinery and environment" (corr. 0.84-0.47; p-value 5.79 × 10-25--3.59 × 10-6). We answered the research question regarding which method, among supervised, unsupervised, or other, is most suitable for application to data in the construction industry.

CONCLUSIONS

The meta-analysis provided an overall estimate of the better effectiveness of supervised methods (Odds Ratio = 0.71, Confidence Interval 0.53-0.96) compared to not-supervised methods.

From unstructured accident reports to a hybrid decision support system for occupational risk management: The consensus converging approach

INTRODUCTION

Workplace accidents in the petroleum industry can cause catastrophic damage to people, property, and the environment. Earlier studies in this domain indicate that the majority of the accident report information is available in unstructured text format. Conventional techniques for the analysis of accident data are time-consuming and heavily dependent on experts' subject knowledge, experience, and judgment. There is a need to develop a machine learning-based decision support system to analyze the vast amounts of unstructured text data that are frequently overlooked due to a lack of appropriate methodology.

METHOD

To address this gap in the literature, we propose a hybrid methodology that uses improved text-mining techniques combined with an un-bias group decision-making framework to combine the output of objective weights (based on text mining) and subjective weights (based on expert opinion) of risk factors to prioritize them. Based on the contextual word embedding models and term frequencies, we extracted five important clusters of risk factors comprising more than 32 risk sub-factors. A heterogeneous group of experts and employees in the petroleum industry were contacted to obtain their opinions on the extracted risk factors, and the best-worst method was used to convert their opinions to weights.

CONCLUSIONS AND PRACTICAL APPLICATIONS

The applicability of our proposed framework was tested on the data compiled from the accident data released by the petroleum industries in India. Our framework can be extended to accident data from any industry, to reduce analysis time and improve the accuracy in classifying and prioritizing risk factors.

Contextualizing injury severity from occupational accident reports using an optimized deep learning prediction model

Background

This study introduced a novel approach for predicting occupational injury severity by leveraging deep learning-based text classification techniques to analyze unstructured narratives. Unlike conventional methods that rely on structured data, our approach recognizes the richness of information within injury narrative descriptions with the aim of extracting valuable insights for improved occupational injury severity assessment.

Methods

Natural language processing (NLP) techniques were harnessed to preprocess the occupational injury narratives obtained from the US Occupational Safety and Health Administration (OSHA) from January 2015 to June 2023. The methodology involved meticulous preprocessing of textual narratives to standardize text and eliminate noise, followed by the innovative integration of Term Frequency-Inverse Document Frequency (TF-IDF) and Global Vector (GloVe) word embeddings for effective text representation. The proposed predictive model adopts a novel Bidirectional Long Short-Term Memory (Bi-LSTM) architecture and is further refined through model optimization, including random search hyperparameters and in-depth feature importance analysis. The optimized Bi-LSTM model has been compared and validated against other machine learning classifiers which are naïve Bayes, support vector machine, random forest, decision trees, and K-nearest neighbor.

Results

The proposed optimized Bi-LSTM models' superior predictability, boasted an accuracy of 0.95 for hospitalization and 0.98 for amputation cases with faster model processing times. Interestingly, the feature importance analysis revealed predictive keywords related to the causal factors of occupational injuries thereby providing valuable insights to enhance model interpretability.

Conclusion

Our proposed optimized Bi-LSTM model offers safety and health practitioners an effective tool to empower workplace safety proactive measures, thereby contributing to business productivity and sustainability. This study lays the foundation for further exploration of predictive analytics in the occupational safety and health domain.

Occupational Injury Risk Mitigation: Machine Learning Approach and Feature Optimization for Smart Workplace Surveillance

Forecasting the severity of occupational injuries shall be all industries' top priority. The use of machine learning is theoretically valuable to assist the predictive analysis, thus, this study attempts to propose a feature-optimized predictive model for anticipating occupational injury severity. A public database of 66,405 occupational injury records from OSHA is analyzed using five sets of machine learning models: Support Vector Machine, K-Nearest Neighbors, Naïve Bayes, Decision Tree, and Random Forest. For model comparison, Random Forest outperformed other models with higher accuracy and F1-score. Therefore, it highlighted the potential of ensemble learning as a more accurate prediction model in the field of occupational injury. In constructing the model, this study also proposed the feature optimization technique that revealed the three most important features; 'nature of injury', 'type of event', and 'affected body part' in developing model. The accuracy of the Random Forest model was improved by 0.5% or 0.895 and 0.954 for the prediction of hospitalization and amputation, respectively by redeveloping and optimizing the model with hyperparameter tuning. The feature optimization is essential in providing insight knowledge to the Safety and Health Practitioners for future injury corrective and preventive strategies. This study has shown promising potential for smart workplace surveillance.

Application of a Machine Learning-based Decision Support Tool to Improve an Injury Surveillance System Workflow

Abstract Background Emergency department (ED)-based injury surveillance systems across many countries face resourcing challenges related to manual validation and coding of data. Objective This paper describes the evaluation of a machine learning-based Decision Support Tool (DST) to assist injury surveillance departments in the validation, coding and use of their data, comparing outcomes in coding time and accuracy pre- and post-implementation. Methods Manually coded injury surveillance data has been used to develop, train and iteratively refine a machine learning-based classifier to enable semi-automated coding of injury narrative data. This paper describes a trial implementation of the machine learning-based DST in the Queensland Injury Surveillance Unit (QISU) workflow using a major pediatric hospital's emergency department data comparing outcomes in coding time and accuracy pre- and post-implementation. Results The study found a 10% reduction in manual coding time after the DST was introduced. The Kappa statistics analysis in both DST-assisted and unassisted data shows increases in accuracy across three data fields; injury intent (85.4% unassisted vs. 94.5% assisted), external cause (88.8% unassisted vs. 91.8% assisted) and injury factor (89.3% unassisted vs. 92.9% assisted). The classifier was also used to produce a timely report monitoring injury patterns during the COVID-19 pandemic. Hence, it has the potential for near real-time surveillance of emerging hazards to inform public health responses. Conclusions The integration of the DST into the injury surveillance workflow shows benefits as it facilitates timely reporting and acts as a DST in the manual coding process.