A text mining-based thematic model for analyzing construction and demolition waste management studies

Area of Science:

• Environmental Science
• Civil Engineering
• Data Science

Background:

• Construction and demolition waste (CDW) management is a significant global challenge.
• The vast body of literature on CDW management poses difficulties for researchers in identifying key trends and insights.
• Automated methods for literature analysis are needed to efficiently extract knowledge from extensive research.

Purpose of the Study:

• To develop and present a comprehensive thematic model for analyzing CDW management literature.
• To automatically detect research insights and map the knowledge domains within CDW management.
• To identify future research directions in the field of CDW management.

Main Methods:

• A comprehensive thematic model was developed by combining Latent Dirichlet Allocation (LDA), word2vec, and a community detection algorithm.
• The model was applied to a dataset of 641 articles on CDW management published between 2000 and 2019, sourced from the Web of Science database.
• Python was utilized for the implementation of the analysis.

Main Results:

• The analysis revealed a four-domain knowledge map in CDW management research.
• The identified domains include: introducing the current situation of CDW management, quantifying CDW generation, assessing CDW and by-products, and facilitating waste diversion.
• The study successfully mined insights from the literature, highlighting key research areas.

Conclusions:

• The comprehensive thematic model is effective for mining insights from large volumes of scientific literature in CDW management.
• The findings provide a structured overview of the CDW management research landscape and suggest avenues for future investigation.
• This approach offers a valuable tool for researchers seeking to understand and contribute to the field of CDW management.