Discovering the Research Topics on Construction Safety and Health Using Semi-Supervised Topic Modeling

Abstract

Safety and health have been one of the major issues in the construction industry worldwide for decades, and the relevant research has correspondingly drawn much attention in the academic field. Considering the expanding size and increasing heterogeneity of this research field, this paper proposes the topic modeling approach to cluster latent topics, extract coherent keywords, and discover evolving trends over the past three decades. Focusing on a total of 1984 articles published in 27 different journal sources until February 2023, this paper applied both unsupervised topic modeling techniques—Latent Dirichlet Allocation (LDA) and Correlation Explanation (CorEx)—and their semi-supervised versions—Guided LDA and Anchored CorEx. The evolving trends and inter-relationship of 15 research topics generated by the Anchored CorEx model (the best-performing model) were analyzed. Top-listed documents of major topics were analyzed to discuss their standalone research focuses. The results of this paper provided helpful insights and implications of existing research and offered potential guides for future research on construction safety and health by helping researchers (1) select research topics of interest and clearing decaying topics; (2) extract the top words of each research topic using systematic approaches; and (3) explore the interconnection of different research topics as well as their standalone focuses.

Structural Equation Modeling in Technology Adoption and Use in the Construction Industry: A Scientometric Analysis and Qualitative Review

“…Keywords are short phrases that represent the core content of the article. Keyword analysis is crucial because scientific publications contain valuable textual information that can represent the primary interests and hotspots of specific fields [51]. This analysis takes into account both Author Keywords (keywords that are manually entered by authors) and Keywords Plus (keywords that are automatically suggested by programs).…”

Integrating Information Entropy and Latent Dirichlet Allocation Models for Analysis of Safety Accidents in the Construction Industry

“…For instance, Liu et al employed text mining in risk assessment to construct knowledge graphs from accident reports [21], while Zhang et al used text mining and natural language processing (NLP) to conduct a comprehensive study of construction safety accident investigation reports [9], and the TF-IDF method was used to optimize the weights and extract the main factors that led to the accidents. Zhou et al utilized the LDA model and related explanatory clustering to identify and analyze the significant areas of research in the field of construction safety and health [22]. Zhong et al [23] proposed an intelligent text mining method combining deep learning and LDA modeling to test the effectiveness of the method with the text of accidents occurring at construction sites on the OSHA website.…”

Identifying Critical Factors and Trends Leading to Fatal Accidents in Small-Scale Construction Sites in Korea

“…In response to the importance of diverse data sourcing, it is worth mentioning that while our primary dataset originated from official sources, the inclusion of data from broader platforms, such as social media enterprises, could potentially enhance the comprehensiveness of the analysis. Such datasets, although rich in capturing real-time sentiments, can pose challenges related to data validity, consistency, and potential biases [18].…”