All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
Based on the reassessment of one of the original reviewers and my own checks, I recommend this paper now for acceptance as the last round of minor revisions is satisfactory.
[# PeerJ Staff Note - this decision was reviewed and approved by Shawn Gomez, a PeerJ Section Editor covering this Section #]
It seems that the authors have amended the manuscript according to the suggestion. Hence, it can be accepted for publication.
It seems good
It is up to the mark
Please address and/or provide responses to the minor revisions suggested by R1 before the paper can be considered ready for publication.
The manuscript meets the standards of clarity and professional English overall, with no major issues in language or presentation. Expanding the introduction to emphasize the gaps in existing research and how this study fills those gaps would make the motivation clearer to readers. A brief mention of challenges or limitations in previous approaches could also strengthen the argument for why NLP techniques were used in this study.
Experimental design is well-conceived, but further elaboration on the technical aspects, including preprocessing steps, evaluation methods, and computing infrastructure, would improve transparency and replicability..
I appreciate adding a section on the stop-word removal. However, explaining why certain stop words or numerical data were removed, or whether any stemming or lemmatization was performed, would help readers understand the process better.
The study’s novelty lies in applying advanced NLP techniques to analyze the intersection of sustainability and circular economy in construction. However, the broader impact of these findings on both academic research and industry practices could be more explicitly addressed. The study offers a systematic comparison between "Circular Construction" (CC) and "Sustainable Construction" (SC), but the long-term implications of these findings for construction practices, policy-making, and future research could be elaborated further.
The conclusion is well-written, but it could benefit from a more detailed discussion on the limitations of the study and potential future research avenues. While the article does mention the temporal scope limitation (focusing on publications from 2021-2024), a short discussion on the possible changes in the outcomes with changing the timeframe to a longer one should be discussed.
I still think that the file "requirements.txt" is needed for full reproductibility, because mentioned packages, do not state their versions, so actually trying to reproduce might be impossible. Environment details,
including especially the hardware specifications is not enough in this case.
It seems that authors have addressed all the comments. Hence, the paper can be accepted for publication.
Sounds good.
Sounds good.
It can be accepted for publication
Please respond in detail to all the comments from the reviewers.
- The article uses professional language and adheres to academic standards, but there are occasional ambiguities and instances where clarity could be improved. For example: Some technical terms (e.g., "Concept Matrix," "TextRank") are used without sufficient explanation early on, which may confuse readers unfamiliar with NLP methodologies.
Suggestions:
Provide clear, concise definitions of technical terms at first mention to improve accessibility.
Simplify overly dense sentences for better readability.
- Literature is well referenced and relevant. The introduction provides a good context for the study, outlining the relevance of sustainability and circular economy (CE) in construction. Relevant literature is cited, but some foundational references appear underrepresented, particularly those on:
Broader discussions of NLP methodologies in sustainability research. Is there such discussion, or is this research new?
Suggestions:
Expand the introduction to include a brief discussion of foundational NLP methodologies and their relevance to this study.
- The structure generally conforms to PeerJ standards and discipline norms. The organization of results into distinct methodological sections enhances clarity.
- The introduction sufficiently introduces the subject but could better emphasize the unique contribution of this study, particularly how it bridges sustainability and CE in construction through NLP.
- Formal results should include clear definitions of all terms and theorems, and detailed proofs (where necessary). Provide clear definitions for key terms (e.g., "Circular Construction," "Sustainable Construction") at their first mention.
- The article fits within the Aims and Scope of PeerJ Computer Science. It focuses on applying natural language processing (NLP) techniques—an area firmly within computer science—to analyze sustainability and circular economy (CE) concepts in construction.
- The study employs recognized NLP techniques (TF-IDF, TextRank, Concept Matrix) to analyze a substantial dataset of scientific articles. However, some limitations in the rigor arise from the narrow temporal scope (2021–2024), which could introduce bias.
- The methods are well-documented, but could benefit from providing computing environment details, including especially the hardware specifications. I have also did not find the file "requirements.txt" which is mentioned in the file "cs-105193-Reproducibility.pdf".
- The article discusses data preprocessing but does not provide enough detail to ensure replication. For instance, it mentions the removal of stop words and numerical data. The histograms demonstrate the distribution of sentences and words in the articles after preprocessing, but the article would benefit from a comparison made to histograms of the words in the articles before the preprocessing, to better understand the relevance of the process.
- The evaluation methods (e.g., Pearson and Spearman correlations) are appropriate but lack sufficient detail on the rationale for choosing these specific metrics.
- The article cites many relevant sources, but there are possible improvements that would enhance the already well written manuscript. I would suggest supplying some technical aspects of TF-IDF, TextRank, and Concept Matrix, and also putting more emphasis on the citations behind choosing Pearson and Spearman correlations.
- The article focuses on applying NLP techniques to analyze sustainability and circular economy (CE) concepts in construction which is indeed very interesting concept.
- The conclusions are well-articulated but are largely limited to summarizing the findings without expanding on their broader significance. While the study identifies differences between "Circular Construction" (CC) and "Sustainable Construction" (SC) and explains the analytical results, there is little discussion on how these insights can influence the design or implementation of sustainability policies in construction.
I would suggest to expand the conclusions to include a discussion on how these findings could influence the construction industry, policy development, or future academic work.
- The study presents a solid analysis of the concepts, using well-established NLP techniques like TF-IDF, TextRank, and Concept Matrix. However, as mentioned previously, there are some limitations related to the narrow scope of the dataset, especially the time related constraint.
- The study generally follows through on the goals set out in the Introduction. The authors aim to explore the relationship between sustainability and CE in the construction sector and use NLP techniques to uncover key terms and their associations. While the analysis and findings are logically presented, the argument could be more robust if the authors had better clarified how the distinction between CC and SC can impact real-world construction practices.
- The conclusion mentions some limitations, such as the narrow temporal scope of the dataset, but does not go into enough depth regarding the potential for further research. It would be beneficial to elaborate on future directions, such as how NLP methods can be applied to further domains within construction, or how the distinctions between CC and SC could be explored in more depth through case studies or longitudinal research.
- I would like to express my appreciation for the concept behind this article, as it addresses an important and emerging area of study in both sustainability and circular economy (CE) within the construction industry. The use of natural language processing (NLP) methods like TF-IDF, TextRank, and Concept Matrix to analyze terms and concepts within the field of sustainability and CE is innovative, as it brings a data-driven, analytical approach to understanding complex, interdisciplinary topics. This study has great potential to contribute to the academic discourse on sustainability practices in construction and to facilitate clearer communication among professionals in the field.
- The article's exploration of the distinctions and similarities between "Circular Construction" (CC) and "Sustainable Construction" (SC) is particularly interesting. By identifying operational aspects like resource efficiency and waste management in CC, and the broader, more holistic approach of SC, this paper offers valuable insights that could guide both industry practitioners and policymakers in refining their strategies for sustainable construction.
- Additionally, I find it noteworthy that the study uses NLP techniques to uncover associations between sustainability concepts and construction-related terminology. For example, the finding that sustainability is more deeply integrated into urban planning practices than circular economy principles suggests a potential gap that could be explored further. This kind of insight, generated through computational analysis, adds an innovative layer to understanding the relationship between theory and practice in these fields.
- Overall, I think this study is on a promising path, and the methodology employed demonstrates a novel approach to examining sustainability and circular economy in construction. The integration of machine learning and NLP with sustainability research is an exciting direction for future studies in this area. I look forward to seeing how the authors can refine and expand their work.
Please check the enclosed file
Please check the enclosed file
Please check the enclosed file
The author shall highlight the practical significance of the study.
How the findings of the study help future researchers working on sustainable construction and circular construction areas?
Why are only the last 3 years (2021-2024) included in the study?
Why are only review papers included? Justify this selection criteria.
Figure 3 is referred to earlier than Figure 2 in the text
Numerical results shall be presented for TF-IDF and Text Rank. The manuscript shall further explain how the TF-IDF and Text Rank results determined the key concepts and themes.
Results derived from the three methods are different. How did the authors conclude the study? Similarities and dissimilarities among the three shall be represented in a tabular format. A discussion should be added on the dissimilarities of the results.
Don’t use abbreviations in the Abstract.
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.