Deep Learning and Text Mining: Classifying and Extracting Key Information from Construction Accident Narratives

Abstract

Construction accidents can lead to serious consequences. To reduce the occurrence of such accidents and strengthen the execution capabilities in on-site safety management, managers must analyze accident report texts in depth and extract valuable information from them. However, accident report texts are usually presented in unstructured or semi-structured forms; analyzing these texts manually requires a lot of time and effort, it is difficult to cope with the demand of analyzing a large number of accident texts, and the quality of key information extracted manually may be poor. Therefore, this study proposes a classification method based on natural language processing (NLP) technology. First, we developed a text classification model based on a convolutional neural network (CNN) that can automatically classify accident categories based on accident text features. Next, taking the classified fall accidents as an example, we extracted key information from accident narratives using the term frequency-inverse document frequency (TF-IDF) method and presented it visually using word clouds. The results show that the overall accuracy of the CNN model reaches 84%, which is better than the other three shallow machine-learning models. Then, eight key accident areas and three accident-prone operations were identified using the TF-IDF algorithm. This study can provide important guidance for project managers and can be used for on-site safety management to help prevent production safety accidents.

Harnessing Generative Pre-Trained Transformers for Construction Accident Prediction with Saliency Visualization

“…The "others" includes the accident types (e.g., overexertion, occupational diseases, and fire) that cannot be classified in the remaining five accident types and takes small proportions, which can decrease the reliability of the classification results. The selection of the six types of accidents in this study is based on a comprehensive analysis of accident reports, existing literature, industry standards, and common occurrences in construction sites [13,29,43]. These six accident types encompass a broad spectrum of incidents that are frequently reported and have significant implications for construction site safety.…”

“…These reports encompass event descriptions, timing information, and location details [11,12]. However, analyzing accident report data is often laborious and time-consuming, demanding profound understanding of safety to extract meaningful insights [11,13]. The conventional approach involves the manual classification of accident cases, typically undertaken by safety professionals [11,14].…”

Development of a Site Information Classification Model and a Similar-Site Accident Retrieval Model for Construction Using the KLUE-BERT Model

“…Luo and Hirogane [42] used Doc2Vec and BERT to analyze the similarity between accident cases from a total of 941 construction accidents reported to the Ministry of Health, Labor, and Welfare of Japan. Li and Wu [43] and Luo et al [44] proposed NLP-based classification methods to extract important information by analyzing text from accident reports. Both studies developed models to classify text based on CNN models.…”

An Artificial-Intelligence-Driven Spanish Poetry Classification Framework

“…Torres et al [32] applied five well-known machine learning classifiers for identifying renal complications and hypertensive disorders in a clinical record that was written in Spanish. Li et al [33] proposed a categorization method based on natural language processing (NLP) techniques for analyzing construction accident report texts. The technique is based on convolutional neural networks and can automatically classify accident categories based on accident text features.…”