Normalization of Web of Science Institution Names Based on Deep Learning

Autor: Zijie Jia, Zhijian Fang, Huaxiong Zhang
Jazyk: angličtina
Rok vydání: 2024
Předmět:
Zdroj: Algorithms, Vol 17, Iss 7, p 312 (2024)
Druh dokumentu: article
ISSN: 1999-4893
DOI: 10.3390/a17070312
Popis: Academic evaluation is a process of assessing and measuring researchers, institutions, or disciplinary fields. Its goal is to evaluate their contributions and impact in the academic community, as well as to determine their reputation and status within specific disciplinary domains. Web of Science (WOS), being the most renowned global academic citation database, provides crucial data for academic evaluation. However, due to factors such as institutional changes, translation discrepancies, transcription errors in databases, and authors’ individual writing habits, there exist ambiguities in the institution names recorded in the WOS literature, which in turn affect the scientific evaluation of researchers and institutions. To address the issue of data reliability in academic evaluation, this paper proposes a WOS institution name synonym recognition framework that integrates multi-granular embeddings and multi-contextual information.
Databáze: Directory of Open Access Journals
Nepřihlášeným uživatelům se plný text nezobrazuje