Despite their usefulness in analyzing the uses of research papers within and beyond academia, the availability of download data remains scarce. Using data from the Érudit journal platform, this paper analyzes the effect of open access availability on numbers of downloads and their geographic diversity. Results show that OA positively affects the numbers of downloads by a factor of 3.0 during the embargo period, and that downloads may maintain this advantage during the following years. Eighty percent of downloads originated from non-subscribing visitors, with the end of the embargo period being strongly associated with an expansion of their geographic diversity. On the whole, our results show that journals can substantially broaden international readership by adopting immediate OA, and that metadata quality is at least as influential as OA for downloads.
Isla, Malena Méndez; Kozlowski, Diego; Larivière, Vincent
Artificial intelligence (AI) is reshaping scientific practice. While recent advances in large language models (LLMs) are substantial, machine learning (ML) has a long history of developments across science. This study analyses the evolution of machine learning in global science from 1990 to 2025, by examining 4.9 million articles from OpenAlex. Using a hierarchical taxonomy of 252 ML techniques and architectures, we found that while deep learning exploded post-2015 and LLMs after 2022, established methods, such as linear and logistic regressions, declined in their share of the total scientific output. Utilizing state-of-the-art embeddings (mE5) and dimensionality reduction (UMAP), we constructed the ML Semantic Space that reveals a core-periphery structure dominated by the Physical Sciences. Computer Science acts as the primary methodological anchor, while Engineering drives applied developments. Health Sciences, led by Medicine, represents the second-largest expansion zone of ML techniques.
Sott, Michele Kremer; Mikhaĭlov, A. A.; Ruffoni, Janaína; Tello‐Gamarra, Jorge; Fischer, Bruno Brandão
Numerous studies have examined innovative partnerships between universities and industrial companies; however, universities are typically treated as a monolithic unit of analysis. Shifting the focus to individual academic research groups (RGs) provides a better understanding of the complexity and heterogeneity of such interactions. This article aims to analyze the interactions between Brazilian industrial firms and university research groups to understand how specific characteristics of these micro-level units influence firms' innovative outcomes. The research utilizes data from the Brazilian national census of research groups (DGP-CNPq) combined with a survey of industrial firms that collaborated with these groups. The study develops five econometric models to measure the effects of firm and research group characteristics on the dimensions of innovation performance, such as technological and business capabilities, and product and process innovation. The results indicate that research group characteristics do not directly affect a firm’s innovation performance. Instead, the presence of a robust knowledge infrastructure and involvement in R&D departments are the primary determinants of enhanced technological and business capabilities in collaborative agreements. Furthermore, while applied science is a significant predictor for product innovation, process innovation is primarily driven by the firm's own technological intensity and the university’s knowledge infrastructure. This study is among the first to provide a granular, micro-level analysis of how research group characteristics — rather than university-wide traits — shape the tangible benefits firms derive from collaboration, specifically within the context of an emerging economy.
Nazarovets, Maryna; Beigel, Fernanda; Van Schalkwyk, François B.; Taşkın, Zehra; Meng, Ping
University journals (UJs) are often described as local or inward-looking, although national and institutional embeddedness are rarely distinguished. This study examines the internationally indexed segment of university journals, analysing Web of Science-indexed titles from seven countries with contrasting publishing traditions and evaluation regimes (Argentina, China, Germany, South Africa, Türkiye, Ukraine, and the United Kingdom), based on publications from 2022-2024. National anchoring (N-Anch) measures the share of papers with at least one author from the journal’s home country, while institutional anchoring (I-Anch) captures participation by authors from the publishing university. The results reveal a clear divergence between these dimensions. National anchoring is high in Türkiye, China, Ukraine, South Africa, and Argentina, but considerably lower in Germany and UK university-press journals. Institutional anchoring remains low across all countries. Thus, strong integration into a national research community does not imply institutional anchoring. The findings show that patterns of national and institutional anchoring vary substantially across countries with different research, evaluation, and publishing systems and cannot be explained by university ownership alone.
English functions as the principal shared language of international scholarly communication, but its dominance also creates inequalities in the production, dissemination, visibility, and assessment of research. Developed as a complementary output of the CoARA Working Group on Multilingualism and Language Biases in Research Assessment, this report examines the historical development of English dominance and the contemporary mechanisms through which it is reproduced. It combines evidence from the literature with descriptive bibliometric analyses based on Web of Science, Scopus, OpenAlex, and CWTS Journal Indicators. The findings show that English is especially dominant in journals with high field normalised citation impact and in the selective segments of scholarly communication most frequently used in research evaluation. However, the extent of this dominance varies considerably across disciplines and countries, and publication in national languages remains important in the social sciences, humanities, and fields directed toward local professional and societal audiences. The report also demonstrates how database selection, metadata quality, and analytical choices affect the visibility of multilingual scholarship. Web of Science and Scopus predominantly represent English language journal literature, whereas OpenAlex reveals a much broader body of multilingual research, although incomplete language, authorship, and institutional metadata remain important limitations. The report further examines the financial, linguistic, and time burdens associated with English language publication and considers how publishing infrastructures, bibliometric indicators, rankings, and assessment practices influence which research becomes visible and valued. This report contributes to the landscape analysis conducted by the Coalition for Advancing Research Assessment (CoARA) Working Group (WG) on Multilingualism and language biases in research assessment and it complements the Implementation proposal for language-aware assessments (Annex II.3).
Damian, Ieda Pelógia Martins; Araujo de Rezende Alvares, Lillian Maria; Segundo, Washington; Garcia de Souza, Marcel
Introduction: technological information plays a strategic role in scientific, technological, and industrial development, especially in contexts marked by innovation and competitiveness. Objective: this study aims to present an updated concept of technological information that reflects current demands in academic and productive sectors. Methodology: a Systematic Literature Review (SLR) was conducted using national and international databases in the field of Information Science. The search resulted in 1,108 documents, from which 68 relevant definitions of technological information were selected and analyzed. Results: the analysis enabled the identification of recurring themes and common elements in existing definitions, allowing for a deeper understanding of the concept. The study highlights the main characteristics and functions of technological information, such as its association with innovation, competitiveness, and strategic decision-making. It emphasizes its role in improving products, processes, and services, and its relevance in patents and research outputs. Conclusion: the study proposes a new definition of technological information as knowledge related to technologies that generate strategic value through their application. This updated definition contributes to theoretical advances in Information Science and offers practical support to organizations aiming to use technological information as a strategic asset. It also broadens the understanding of the term’s scope, reinforcing its importance for sustainable development and innovation.
Lorensi do Canto, Fábio; Carneiro, Renan; Dias, Thiago; Talau, Marcos; Segundo, Washington
This dataset provides a structured JSON representation of the International Patent Classification (IPC) scheme used in the prototype described in the associated paper. The dataset encodes the hierarchical organization of IPC symbols, supporting navigation across classification levels and the construction of Boolean search strategies for exact and hierarchical patent retrieval. The JSON structure was prepared to support a web-based patent search prototype that enables bilingual IPC navigation in English and Portuguese and assists users in formulating IPC-based queries for execution in the beta version of the new patent search system of the Brazilian Industrial Property Office (INPI). By making the IPC scheme available as a reusable dataset, this deposit supports transparency, reproducibility, and reuse of the classification structure employed in the study. This dataset is intended as a research component associated with the paper and may be used for studies or systems involving patent information retrieval, IPC-based query expansion, hierarchical classification browsing, and the development of search interfaces based on patent classification schemes.
Local research has gained increasing attention in recent years, with a growing body of literature emphasizing its significance—especially for peripheral communities underrepresented in mainstream scientific discourse. However, despite this growing interest, there remains no clear consensus on how to define local research. Conceptual clarity is essential for conducting consistent and replicable analyses that capture the specific characteristics, contexts, and societal relevance of local research. This need is particularly acute in the Global South, where structural inequalities in visibility and recognition often marginalize regionally focused scholarship. In Africa, for instance, a substantial share of research is published in local or regional journals that are poorly represented in global indexing databases. As bibliometric analyses typically rely on these databases, this exclusion diminishes their perceived relevance. To address this gap, the present study incorporates data from regional and open-access sources to examine citation patterns associated with African local research. It uses multiple conceptualizations of local research to classify publications as either local or non-local depending on the definition used. These classifications are correlated with three citation-based indicators: average citations per publication, time to first citation, and the proportion of uncited publications. Additionally, the study explores the characteristics of citing publications, distinguishing between local and non-local sources as well as domestic and international citations. The findings reveal that citation patterns differ significantly across locality definitions, underscoring the importance of adopting nuanced, context-sensitive approaches to assess and support local research.
Segundo, Washington; Dias, Thiago Magela; Souza, Marcel Garcia; do Canto, Fábio Lorensi; Sena, Priscila Machado
The adoption of the FAIR principles (Findable, Accessible, Interoperable, Reusable) has become a cornerstone of Open Science, guiding practices aimed at improving the quality, transparency, and reuse of scientific data. In the context of CRIS (Current Research Information Systems), these principles play a strategic role in supporting the integration, standardization, and governance of scientific information on a national scale. This article analyzes the application of the FAIR principles within the Brazilian Scientific Research Information Ecosystem (BrCris), developed by the Brazilian Institute of Information in Science and Technology (Ibict), with the aim of discussing how its technical and organizational practices contribute to building a more reliable, interoperable, and reusable scientific data infrastructure. The methodology adopted is based on a literature review of the FAIR principles and a document analysis of technical reports, scientific articles, and institutional materials related to BrCris. The results indicate that BrCris shows significant progress in adopting the FAIR principles, especially in the use of persistent identifiers, semantic data integration, interoperability with international infrastructures, and the availability of data in open and reusable formats. It is concluded that BrCris constitutes a strategic infrastructure for the implementation of the FAIR principles in Brazil.
Bin, Adriana; Pinto, Daniela Maciel; Pereira, Cesar A. Guimarães; Cristofoletti, Evandro Coggo; Mena-Chalco, Jesus Paschoal; Salles Filho, Sérgio; Sugimoto, Cassidy R.
Peer review plays a central role in research funding decisions, yet the evaluative language used in review reports remains insufficiently explored, particularly in funding agencies. This study examines how linguistic and sentiment characteristics of peer-review reports relate to funding outcomes at the São Paulo Research Foundation (FAPESP), Brazil, focusing on the Regular Research Grant (RRG) and Young Investigator Grant (YIG) schemes between 2000 and 2017. Drawing on 19,333 proposals and their corresponding review reports, we analyze three aggregated corpora — overall reports, proposal-related reports, and applicant track-record reports — using natural language processing and sentiment analysis. We compare approved and rejected proposals across fields of knowledge, reviewer gender, applicant gender, and peer-review report sections. Results show that reviewers use distinct textual and sentiment patterns when justifying approval or rejection. These patterns vary according to the section evaluated — applicant track record or technical proposal — and also differ by field of knowledge and, more locally, by reviewer–applicant gender configurations. The pattern is more regular and stable in RRG than in YIG. Overall, the findings show how evaluative language illuminates fairness, consistency, and transparency in grant peer review.
While most research funding is national, international funding plays a relatively important role, particularly for some regions of the Global South. However, the extent of funding flows between world regions remains unclear. This study represents a first attempt to map global research funding flows based on a set of grant metadata from Dimensions, augmented with data from the Gates Foundation and the European Commission (CORDIS). We find major regional gaps in the data outside of the leading scientific regions. With the data currently available, we find that Europe and Sub-Saharan Africa are the biggest recipients of inter-regional funding, with North America as main funder, and that over the last 25 years, total annual research funding has grown to above USD 60 billion, while inter-regional funding flows have stagnated at around USD 1.6 billion per year. These insights are contingent on data availability, raising the urgency for improving funding metadata infrastructures.
Funding acknowledgments (FAs) provide a potential but underexplored source for mapping global research funding flows. This study develops a framework for constructing global funding flow measures using FA data from Dimensions and assesses both the opportunities and limitations of this approach. We compare total and inter-regional funding flows based on funding grant information available in FAs of scientific publications. Our analysis identifies a key limitation arising from collaborative publications, where funding is attributed at the publication level rather than to individual authors. This obscures the interpretation of cross-border funding flows. We show that this attribution structure introduces systematic ambiguity and limits the validity of FA-based analyses. We argue that disaggregated, author-level funding attribution is essential for accurately tracing funding flows. This requires coordinated action among funders, publishers, and researchers, as well as the adoption of more standardized, machine-readable FAs formats.
This is a position paper (still as a work-in-progress) with the main insights of the project "Tracking Research Funding Flows in the Global South" funded by IDRC and presented in the Wellcome Trust in the workshop 'Mapping research funding flows for global alignment and impact' on 23rd of April 2026. The full list of contributors to the project is included in the document. A long and a short presentation of this position paper are also attached as powerpoint files. A recording of the long presentation in the centre CENIT of the Universidad San Martín de Buenos Aires, in Spanish, is available here.
area, CWTS Engagement & Inclusion focal; de Bordes, Carole; Costas, Rodrigo; Holtrop, Tjitske; Traag, Vincent
This document outlines the strategic plan for the Engagement & Inclusion focal area at the Centre for Science and Technology Studies (CWTS) at Leiden University for the period 2026–2028, presenting the focal area’s vision, long-term strategic impacts, and operational objectives developed through collaborative discussions among members and coordinators.
Cavasin, Mariana Luiza Castilho; Carneiro, Ana Maria
Funded by CAPES
In recent decades, countries of origin have increasingly sought to develop policies to engage with their scientific diaspora. These policy designs typically vary across models of repatriation, linkage, and circulation, aiming to strengthen national science, technology and innovation (STI) systems. This article analyses the design and early implementation of the Knowledge Brazil Program to identify the strategies prioritized by the Brazilian state in engaging with its scientific diaspora. The study is based on document analysis, program microdata, and media sources, including news reports and opinion articles from widely circulated outlets. The findings are synthesized through a SWOT analysis, identifying strengths in the program's design, which incorporates the three core components of diaspora engagement—repatriation, linkage, and circulation; weaknesses in the asymmetry between the program's axes, with a clear emphasis on repatriation; opportunities in the diaspora's willingness to contribute to the national STI system; and threats related to concerns regarding the absorption of researchers into academia after the program's completion.
Carneiro, Ana Maria; Gimenez, Ana Maria Nunes; Capocchi, Eduardo Rodrigues; Bezerra, Luiza Maria Capanema; Ferrero, Luciane Graziele Pereira
Funded by UNICAMP
Academic mobility is a central issue in the study of academic professions, shaping career trajectories, institutional dynamics, and the distribution of knowledge production. While international research has explored its determinants and implications, evidence from Global South contexts remains scarce. Brazil, with one of the largest higher education systems in the world, provides a valuable case to examine how institutional, sociodemographic, and policy factors shape academic mobility. This study investigates the employment trajectories of professors affiliated with Brazilian higher education institutions between 2003 and 2021, based on microdata from the Annual Social Information Report (RAIS). The analysis accounts for gender, age, degree, institutional type, distance, and remuneration. Nearly half of faculty (46.9%) experienced at least one mobility event, but mobility is highly segmented: it is concentrated in teaching-intensive private institutions, while research-intensive and graduate-oriented public universities exhibit markedly lower mobility and stronger retention. Most moves occur within the same institutional segment, suggesting barriers to entry and circular mobility within less prestigious tiers. Spatial mobility is likewise concentrated: 62.6% of moves occur within 0–50 km and 28.1% within 50–500 km, with evidence of greater geographic anchoring later in the career. Salary effects are asymmetric by direction: transitions from public to private institutions are associated with short-term wage losses, whereas moves from private to public institutions show sizable gains that materialize more gradually. The study contributes original evidence to the international literature on academic mobility, while also highlighting data limitations in Brazil. It calls for complementary qualitative research and deeper exploration of labor precarization in the private sector.
Rafols, Ismael; Costas, Rodrigo; Bezuidenhout, Louise; Brasil, André
For science and technology to contribute to social justice, scientometrics analyses need to be able to produce descriptions that are appropriate to specific contexts and values. Given that the participation of stakeholders in the analyses is crucial for these perspectives to be realized, this requires research information analyses that are “open”. In this paper, we propose that this “openness” in research information concerns several dimensions. In the first place, research information should be open in the sense of being transparent and accessible. Second, research information should be open in the sense of being inclusive and diverse, which includes two dimensions: that stakeholders can actually use it, and that the contents include different types of knowledge. Third, the research information should be provided in forms that empower interrogation by stakeholders, so that they can retrieve and construct the descriptions of science and technology that are more appropriate for their context. This last step implies efforts to make visible many scientific contributions from the global south which are currently invisible. We call this vision “the multiversatory”: an approach to the observation of science that facilitates pluralistic analyses of the knowledge created across a variety of places and contexts, the pluriverse.
