[Purpose/Significance] To address the bottleneck that a single data source cannot simultaneously ensure wide literature coverage, high-quality entity disambiguation, and strong timeliness, this study aims to construct a high-quality academic knowledge graph that meets more comprehensive needs. The goal is to enhance the reliability of bibliometric analysis and research on academic social networks. [Result/Conclusion] We propose a large language model-assisted framework for constructing a multi-source integrated literature knowledge graph, leveraging the complementary strengths of four mainstream databases: Scopus, Web of Science (WoS), PubMed, and OpenAlex. The framework consists of two major modules: (1) multi-source literature integration and (2) institutional and author entity disambiguation. By harnessing the zero-shot capabilities of large models, the framework ensures high-quality knowledge graph construction. [Result/Conclusion] Using pneumonia research as a case study, we build a multi-source integrated academic knowledge graph encompassing approximately 2.25 million publications. The graph achieves effective disambiguation and annotation of millions of institutional affiliations and tens of millions of author-publication records. Compared with single-source datasets, the integrated graph, while maintaining low redundancy and minimal false merges, significantly improves literature coverage, metadata completeness, and the precision and recall of entity annotations. It also compensates for the shortcomings of individual databases in institutional and author annotations. Moreover, the integrated graph enables reverse correction of source data, thereby enhancing the accuracy of author and institutional network analyses in classical bibliometric tools such as VOSviewer and CiteSpace. It also supports the exploration of innovative bibliometric methods, demonstrating substantial application value. The proposed data integration framework thus provides a feasible pathway and practical reference for constructing high-quality academic knowledge graphs across different disciplines.
[1] ALAM N, JIA L, CHENG A, et al. Global research trends on gut microbiota and metabolic dysfunction-associated steatohepatitis: insights from bibliometric and scientometric analysis[J]. Frontiers in pharmacology, 2024, 15: 1390483.
[2] 王芳. 水环境抗生素研究进展与趋势—— 基于VOSviewer的可视化分析[J]. 应用化工, 2025(11): 2963-2970. (WANG F. Research progress and trend of antibiotics in water environment - visual analysis based on VOSviewer[J]. Applied chemical industry, 2025(11): 2963-2970.)
[3] CAI Y, LI G, WEN L, et al. Intellectual landscape and emerging trends of big data research in hospitality and tourism: a scientometric analysis[J]. International journal of hospitality management, 2024, 117: 103633.
[4] 李燕, 李克建, 王雪梅. 海外中华文明研究与展望—— 基于 Web of Science数据库的分析[J]. 西南民族大学学报(人文社会科学版), 2025, 46(2): 231-240. (LI Y, LI K J, WANG X M. Overseas studies and future prospects of Chinese civilization: an analysis using the Web of Science database[J]. Journal of Southwest Minzu University(humanities and social sciences edition), 2025, 46(2): 231-240.)
[5] WANG K, SHEN Z, HUANG C, et al Microsoft academic graph: when experts are not enough[J]. Quantitative science studies, 2020, 1(1): 396-413.
[6] SCHEIDSTEGER T, HAUNSCHILD R. Which of the metadata with relevance for bibliometrics are the same and which are different when switching from Microsoft Academic Graph to OpenAlex?[J]. Profesional de la información, 2023, 32(2): 1699- 2407.
[7] ZHANG F, LIU X, TANG J, et al. OAG: linking entities across large-scale heterogeneous knowledge graphs[J]. IEEE transactions on knowledge and data engineering, 2023, 35(9): 9225-9239.
[8] VELEZ-ESTEVEZ A, PEREZ I J, GARCÍA-SÁNCHEZ P, et al. New trends in bibliometric APIs: a comparative analysis[J]. Information processing & management, 2023, 60(4): 103385.
[9] DONNER P, RIMMERT C, VAN ECK N J. Comparing institutional-level bibliometric research performance indicator values based on different affiliation disambiguation systems[J]. Quantitative science studies, 2020, 1(1): 150–170.
[10] REHS A. A supervised machine learning approach to author disambiguation in the Web of Science[J]. Journal of informetrics, 2021, 15(3): 101166.
[11] SUBBOTIN A, AREF S. Brain drain and brain gain in Russia: analyzing internationalmigration of researchers by discipline using Scopus bibliometric data 1996–2020[J]. Scientometrics, 2021, 126(9): 7875-7900.
[12] CÉSPEDES L, KOZLOWSKI D, PRADIER C, et al. Evaluating the linguistic coverage of OpenAlex: an assessment of metadata accuracy and completeness[J]. Journal of the Association for Information Science and Technology, 2025, 76(6): 884-895.
[13] VISSER M, VAN ECK N J, WALTMAN L. Large-scale comparison of bibliographic data sources: scopus, web of science, dimensions, CrossRef, and microsoft academic[J]. Quantitative science studies, 2021, 2(1): 20-41.
[14] HUG S E, OCHSNER M, BRÄNDLE M P. Citation analysis with microsoft academic[J]. Scientometrics, 2017, 111(1): 371– 378.
[15] MARTÍN-MARTÍN A, THELWALL M, ORDUNA-MALEA E, et al. Google Scholar, Microsoft Academic, Scopus, Dimensions, Web of Science, and OpenCitations’ COCI: a multidisciplinary comparison of coverage via citations[J]. Scientometrics, 2021, 126(1): 871-906.
[16] 陈柏彤, 康宇杰. 基于分布式语义分析的学术创新跨领域演化探析[J]. 图书情报工作, 2024, 68(12): 95-108. (CHEN B T, KANG Y J. Cross-domain evolution of academic innovation based on distributional semantic analysis[J]. Library and information service, 2024, 68(12): 95-108.)
[17] TANG J, ZHANG J, YAO L, et al. ArnetMiner: extraction and mining of academic social networks[C]//Proceedings of the 14th ACM SIGKDD international conference on knowledge discovery and data mining. New York: ACM, 2008: 990-998.
[18] WANG G, SUN Z, HU W, et al. Author name disambiguation based on heterogeneous graph neural network[J]. Plos one, 2025, 20(2): e0310992.
[19] LIU X, WU K, LIU B, et al. HNERec: scientific collaborator recommendation model based on heterogeneous network embedding[J]. Information processing and management, 2023, 60(2): 103253.
[20] NGUYEN B T, DOAN T T N, HUYNH S T, et al. An end-toend named entity recognition platform for vietnamese real estate advertisement posts and analytical applications[J]. IEEE access, 2022, 10: 87681-87697.
[21] SHAO Z, YUAN S, JIN Y, et al. Scholar’s career switch from academia to industry: mining and analysis from AMiner[J]. Big data research, 2024, 36: 100441.
[22] SONG Y, SITU F, ZHU H, et al. To be the prince to wake up sleeping beauty: the rediscovery of the delayed recognition studies[J]. Scientometrics, 2018, 117(1): 9-24.
[23] PRETOLESI D, STANZANI I, RAVERA S, et al. Artificial intelligence and network science as tools to illustrate academic research evolution in interdisciplinary fields: the case of Italian design[J]. Plos one, 2025, 20(1): e0315216.
[24] ARIA M, LE T, CUCCURULLO C, et al. OpenalexR: an R-tool for collecting bibliometric data from OpenAlex[J]. R journal, 2023, 15(4): 167-180.
[25] ALPERIN J P, PORTENOY J, DEMES K, et al. An analysis of the suitability of OpenAlex for bibliometric analyses[J/OL]. 2024[2025-11-10]. https://arxiv.org/abs/2404.17663.
[26] ZHANG L, CAO Z, SHANG Y, et al. Missing institutions in OpenAlex: possible reasons, implications, and solutions[J]. Scientometrics, 2024, 129(10): 5869-5891.
[27] ZHOU H, SUN M. Evaluating authorship disambiguation quality through anomaly analysis on researchers’ career transition[J/OL]. 2024[2025-11-10]. https://arxiv.org/abs/2412.18757.
[28] ANCONA A, CERQUETI R, VAGNANI G. A novel methodology to disambiguate organization names: an application to EU framework programmes data[J]. Scientometrics, 2023, 128(8): 4447-4474.
[29] DONG Y, CHAWLA N V, SWAMI A. Metapath2vec: scalable representation learning for heterogeneous networks[C/OL]//Proceedings of the 23rd ACM SIGKDD International Conference on knowledge discovery and data mining. Halifax NS, Canada: ACM, 2017: 135-144.
