Automatically learning construction injury precursors from text

• Published in: Automation in construction Vol. 118; p. 103145
• Main Authors: Baker, Henrietta, Hallowell, Matthew R., Tixier, Antoine J.-P.
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01.10.2020; Elsevier BV
• Subjects: Artificial intelligence; Artificial neural networks; Construction accidents & safety; Construction industry; Construction safety; Construction site accidents; Deep learning; Identification methods; Injuries; Knowledge extraction; Natural language processing; Precursors; Safety; Support vector machines; Text mining; Deep learning; Text mining; Natural language processing; Construction safety; Knowledge extraction; Artificial intelligence
• ISSN: 0926-5805; 1872-7891
• DOI: 10.1016/j.autcon.2020.103145

In light of the increasing availability of digitally recorded safety reports in the construction industry, it is important to develop methods to exploit these data to improve our understanding of safety incidents and ability to learn from them. In this study, we compare several approaches to automatically learn injury precursors from raw construction accident reports. More precisely, we experiment with two state-of-the-art deep learning architectures for Natural Language Processing (NLP), Convolutional Neural Networks (CNN) and Hierarchical Attention Networks (HAN), and with the established Term Frequency - Inverse Document Frequency representation (TF-IDF)+Support Vector Machine (SVM) approach. For each model, we provide a method to identify (after training) the textual patterns that are, on average, the most predictive of each safety outcome. We show that among those pieces of text, valid injury precursors can be found. The proposed methods can also be used by the user to visualize and understand the models' predictions. •We propose several methods to automatically learn injury precursors from raw construction accident reports.•Precursors are learned by multiple supervised models as a by-product of training them at predicting safety outcomes.•We experiment with two deep learning models (CNN and HAN), and a more traditional machine learning approach (TF-IDF+SVM).•The proposed methods can also be used to visualize and understand the models' predictions.