Beyond buzzwords: NLP reveals common threads in sustainable and circular construction discourse

Shakarim Aubakirov, Alexandr Pak, Iskander Akhmetov, Aidana Tleuken, Hüseyin Atakan Varol, Assel Zh. Akzhalova, Ferhat Karaca

2025

Full text Open access: Gold Metadata (CSL-JSON)
DOI 10.7717/peerj-cs.3085
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

Circular economy and sustainability have both seen rapid growth in academic literature, often leading to ambiguity and the overuse of these terms. This obscures their true objectives and makes it challenging to discern their distinct intentions. Manually analyzing the vast body of recent publications to understand how these concepts connect to environmentally beneficial practices is laborious and time-consuming. This study aims to compare and analyze existing literature on sustainable and circular construction using natural language processing (NLP) techniques to elucidate the similarities and overlaps between these concepts within the construction industry. To achieve this, we employed three NLP methods: (1) TextRank, a graph-based ranking algorithm that extracts key structural relationships between terms in a document; (2) term frequency–inverse document frequency, a statistical measure that identifies the most significant terms based on their frequency and uniqueness within the corpus; and (3) semantic annotation (Wikifier), a method that links text tokens to structured knowledge bases such as Wikipedia for better contextual understanding. These methods are used to analyze a dataset of 480 academic articles focusing on sustainability and circular economy in the construction sector. Our analysis revealed that circular construction is more specific and practical, emphasizing resource efficiency, waste management, and industry-specific processes, targeting the operational aspects of recycling and resource recovery. In contrast, sustainable construction encompasses a broader and more holistic scope, including urban planning, community development, and long-term environmental impacts. This study demonstrates how NLP methods can systematically disentangle closely related frameworks in construction literature, providing a replicable methodological framework for future data-driven investigations. By clarifying the distinctions and overlaps between the terms “circular construction” and “sustainable construction”, our research offers enhanced understanding for policymakers, industry practitioners, and academics aiming to integrate sustainable and circular principles effectively within the construction sector.