{"title":"通过三个固定的维度追踪图书馆和信息科学的演变:图书馆、人和算法","authors":"Renli Wu , Ruiyang Chen , Lu An , Chuanfu Chen","doi":"10.1016/j.joi.2025.101712","DOIUrl":null,"url":null,"abstract":"<div><div>With rapid technological advancements and societal changes, the field of <em>Library and Information Science</em> (LIS) is dramatically evolving. To capture these shifts, we analyzed over 140,000 LIS publications ranging from 1990 to 2023, examining the discipline’s research evolution across three semantic dimensions: <strong><em>library</em></strong>, representing the historical foundation and institutional infrastructure of LIS; <strong><em>people</em></strong>, representing the core interacting participants and human-centered focus of LIS; and <strong><em>algorithm</em></strong>, representing the methodological advancements driven by emerging technologies in LIS. Utilizing Doc2Vec with a multi-label joint training scheme, we created a consistent embedding space for various LIS entities, including research terms, papers, journals, and countries. By mapping these entities onto a unified framework underpinned by three anchored dimensions, we reveal that the publications of the <strong><em>library</em></strong> dimension, dominant since the 1990s, have declined after 2011, reflected in the focus shifts of LIS research, journal clusters, and nations. Concurrently, LIS research has gravitated toward the <strong><em>people</em></strong> dimension, with people-related studies evolving into a more independent branch. The <strong><em>algorithm</em></strong> dimension is rapidly emerging, with journals more closely associated with it exhibiting higher impact factors, and the research centroids of journals and countries are converging toward it. However, <em>algorithm</em>-dominated research is increasingly detached from the other two dimensions, especially the <em>library</em>. Additionally, developed countries prioritize the research related to <em>library</em> and <em>people</em> dimensions, while developing countries exhibit a stronger emphasis on <em>algorithms</em>-focused publications. To ensure robustness, we further validated our results using a recent <em>ModernBERT</em> model fine-tuned for the LIS context. The findings reveal the developmental dynamics and potential fragmentation within LIS, offering insights for scholars, journals, institutions, and policymakers.</div></div>","PeriodicalId":48662,"journal":{"name":"Journal of Informetrics","volume":"19 3","pages":"Article 101712"},"PeriodicalIF":3.5000,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Tracing the evolution of library and information science through three anchored dimensions: Library, people, and algorithm\",\"authors\":\"Renli Wu , Ruiyang Chen , Lu An , Chuanfu Chen\",\"doi\":\"10.1016/j.joi.2025.101712\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>With rapid technological advancements and societal changes, the field of <em>Library and Information Science</em> (LIS) is dramatically evolving. To capture these shifts, we analyzed over 140,000 LIS publications ranging from 1990 to 2023, examining the discipline’s research evolution across three semantic dimensions: <strong><em>library</em></strong>, representing the historical foundation and institutional infrastructure of LIS; <strong><em>people</em></strong>, representing the core interacting participants and human-centered focus of LIS; and <strong><em>algorithm</em></strong>, representing the methodological advancements driven by emerging technologies in LIS. Utilizing Doc2Vec with a multi-label joint training scheme, we created a consistent embedding space for various LIS entities, including research terms, papers, journals, and countries. By mapping these entities onto a unified framework underpinned by three anchored dimensions, we reveal that the publications of the <strong><em>library</em></strong> dimension, dominant since the 1990s, have declined after 2011, reflected in the focus shifts of LIS research, journal clusters, and nations. Concurrently, LIS research has gravitated toward the <strong><em>people</em></strong> dimension, with people-related studies evolving into a more independent branch. The <strong><em>algorithm</em></strong> dimension is rapidly emerging, with journals more closely associated with it exhibiting higher impact factors, and the research centroids of journals and countries are converging toward it. However, <em>algorithm</em>-dominated research is increasingly detached from the other two dimensions, especially the <em>library</em>. Additionally, developed countries prioritize the research related to <em>library</em> and <em>people</em> dimensions, while developing countries exhibit a stronger emphasis on <em>algorithms</em>-focused publications. To ensure robustness, we further validated our results using a recent <em>ModernBERT</em> model fine-tuned for the LIS context. The findings reveal the developmental dynamics and potential fragmentation within LIS, offering insights for scholars, journals, institutions, and policymakers.</div></div>\",\"PeriodicalId\":48662,\"journal\":{\"name\":\"Journal of Informetrics\",\"volume\":\"19 3\",\"pages\":\"Article 101712\"},\"PeriodicalIF\":3.5000,\"publicationDate\":\"2025-08-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Informetrics\",\"FirstCategoryId\":\"91\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S1751157725000768\",\"RegionNum\":2,\"RegionCategory\":\"管理学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Informetrics","FirstCategoryId":"91","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1751157725000768","RegionNum":2,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS","Score":null,"Total":0}
Tracing the evolution of library and information science through three anchored dimensions: Library, people, and algorithm
With rapid technological advancements and societal changes, the field of Library and Information Science (LIS) is dramatically evolving. To capture these shifts, we analyzed over 140,000 LIS publications ranging from 1990 to 2023, examining the discipline’s research evolution across three semantic dimensions: library, representing the historical foundation and institutional infrastructure of LIS; people, representing the core interacting participants and human-centered focus of LIS; and algorithm, representing the methodological advancements driven by emerging technologies in LIS. Utilizing Doc2Vec with a multi-label joint training scheme, we created a consistent embedding space for various LIS entities, including research terms, papers, journals, and countries. By mapping these entities onto a unified framework underpinned by three anchored dimensions, we reveal that the publications of the library dimension, dominant since the 1990s, have declined after 2011, reflected in the focus shifts of LIS research, journal clusters, and nations. Concurrently, LIS research has gravitated toward the people dimension, with people-related studies evolving into a more independent branch. The algorithm dimension is rapidly emerging, with journals more closely associated with it exhibiting higher impact factors, and the research centroids of journals and countries are converging toward it. However, algorithm-dominated research is increasingly detached from the other two dimensions, especially the library. Additionally, developed countries prioritize the research related to library and people dimensions, while developing countries exhibit a stronger emphasis on algorithms-focused publications. To ensure robustness, we further validated our results using a recent ModernBERT model fine-tuned for the LIS context. The findings reveal the developmental dynamics and potential fragmentation within LIS, offering insights for scholars, journals, institutions, and policymakers.
期刊介绍:
Journal of Informetrics (JOI) publishes rigorous high-quality research on quantitative aspects of information science. The main focus of the journal is on topics in bibliometrics, scientometrics, webometrics, patentometrics, altmetrics and research evaluation. Contributions studying informetric problems using methods from other quantitative fields, such as mathematics, statistics, computer science, economics and econometrics, and network science, are especially encouraged. JOI publishes both theoretical and empirical work. In general, case studies, for instance a bibliometric analysis focusing on a specific research field or a specific country, are not considered suitable for publication in JOI, unless they contain innovative methodological elements.