2026-10-15T08:50:43Z https://data.mendeley.com/oai

oai:data.mendeley.com/gwnjzc8yn9.1 2026-10-06T14:15:06Z

ABEJIDE, KOLAWOLE Machine Learning and Finite Element Modelling in Construction Engineering Research: A Scientometric Review Exposing the Pitting-Corrosion Gap in Steel Bridge Girder Assessment Mendeley Data This dataset supports a scientometric review of machine learning (ML) and finite element (FE) modelling research in construction engineering. The review tests whether research on pitting corrosion in steel bridge girders is visible within that literature. It contains the Scopus export used for science mapping, the VOSviewer output files, and the figures derived from them. The Scopus search was run on 23 September 2026 in the title, abstract and keyword fields. It combined the terms pitting corrosion, bridge steel girders, artificial neural network, machine learning and finite element modelling with the construction-industry context. The initial 2,410 records were narrowed by subject area (Engineering, Computer Science, Environmental Science), document type (articles and conference papers), language (English) and a controlled- keyword limit. This left 1,624 records published between 1995 and September 2026 (1,094 journal articles and 530 conference papers) in 634 sources. Each record has 45 Scopus fields, including authors, title, year, source, citation count, DOI, affiliations, abstract, author and index keywords, references, document type and open-access status. The VOSviewer outputs cover five analyses:  keyword co-occurrence (87 of 11,452 keywords, four clusters)  countries (13 of 99)  sources (10 of 634)  authors (14 of 5,156)  most-cited documents (13 of 1,624) 2 Each has a network, overlay (mean publication year) or density visualisation, or a table exported as text. A world map of document counts by country is also included. The dataset can be used to: • reproduce the bibliometric indicators and maps in the paper • rerun the screening with different thresholds • extend the analysis to other databases, or to a corrosion-focused search without the construction-industry term Notes for reuse: The dataset reflects the Scopus index on the search date, and 2026 is a partial year. The search was deliberately built around the construction-industry context, so it is a test of visibility and not a complete census of corroded-steel research. In the export, 147 records have no author keywords and 82 have no DOI. Suggested keywords: scientometric review; VOSviewer; Scopus; machine learning; finite element modelling; artificial neural networks; pitting corrosion; steel bridge girders; residual strength; construction industry. Construction Industry Machine Learning Pitting Corrosion Girder Finite Element Modeling Abiola-Ogedengbe, Diekolola Dataset doi:10.17632/gwnjzc8yn9.1 oai:data.mendeley.com/gwnjzc8yn9.1 Creative Commons Attribution 4.0 International http://creativecommons.org/licenses/by/4.0 https://data.mendeley.com/datasets/gwnjzc8yn9 2026-10-06T14:15:06Z