Open access resources: the efficiency of metadata and services

The development of open science ideas and principles has facilitated the spread of open access resources (OARs), creating a wide range of information retrieval and bibliometric capabilities for analyzing various documents, journals, authors, organizations, research areas, thanks to the increased effectiveness of metadata, the links between resources, the implementation of proprietary classification systems in OARs, and data enrichment using artificial intelligence technologies. Metadata plays a crucial role in enhancing the effectiveness of OARs, as it serves as a means of document management, ensuring their identification, systematization, visibility, search, discovery, and access control in information retrieval systems. However, unified approaches are lacking regarding 1) the representation of metadata in OARs, 2) the development of classification schemes and services for processing search results, and 3) the presentation of search fields. This exacerbates the problems often encountered when working with information resources related to metadata quality (incompleteness, inaccuracy, errors, inconsistencies, subjectivity in subjectification and systematization) and their interoperability during automatic exchange, processing, and use. The study aims to analyze the effectiveness of metadata representation in major scientific OARs and their functionality for document search and processing. To achieve this goal, the first stage involved a content analysis of 10 reviews from 2023–2025. These reviews compared major scientific OARs (Crossref, Dimensions, OpenAlex, Scilit, Semantic Scholar, The Lens, etc.) with international scientometric databases Scopus and Web of Science Core Collection in terms of metadata completeness and quality, confirming the potential of OARs as bibliographic sources for scientific information retrieval. To validate these findings, the second stage involved an analysis of metadata in four selected major scientific OARs (OpenAlex, Dimensions, The Lens, and Semantic Scholar), assessing their service functionality for conducting bibliometric research and visualizing the resulting data. Furthermore, the proprietary or adopted document classification systems within the OARs were examined, along with the specifics of exporting large volumes of metadata. A comparative analysis of OpenAlex, Dimensions, The Lens, and Semantic Scholar revealed differences in field completeness, analytical services, export limitations, as well as typical metadata quality vulnerabilities. Nevertheless, the prospects for the development of these resources for scientific information retrieval and research are acknowledged, considering their continuously evolving functionality and the use of reliable sources for data enrichment, such as Crossref, ORCID, ROR, etc. The presence of international identifiers, the resolution of ambiguities in description, and addressing issues in document indexing when assigning additional metadata fields in OARs will enable their use as effective information resources. This will contribute to forming a more transparent research landscape, enable multiparameter analysis of the development of scientific fields and collaboration, facilitate multifaceted study of research topics, and enhance the visibility of publications in the global information space. It will also increase the citation potential of authors, scientific organizations, and journals through inclusion in a greater number of interconnected resources, provision of additional metadata for publication, and the availability of service capabilities and functionalities for further processing and presentation using visualization tools, knowledge mapping instruments, and others.