Developing a novel algorithm for comparing cluster patterns in networks on journal articles during and after COVID-19: Bibliometric analysis

Cluster analysis is vital in bibliometrics for deciphering large sets of academic data. However, no prior research has employed a cluster-pattern algorithm to assess the similarities and differences between 2 clusters in networks. The study goals are 2-fold: to create a cluster-pattern comparison al...

Full description

Saved in:
Bibliographic Details
Published in:Medicine (Baltimore) 2024-03, Vol.103 (12), p.e37530-e37530
Main Authors: Wu, Alice-Like, Chow, Julie Chi
Format: Article
Language:English
Subjects:
Academies and Institutes
Algorithms
Bibliometrics
Cluster Analysis
COVID-19
Humans
Systematic Review and Meta-Analysis
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Cluster analysis is vital in bibliometrics for deciphering large sets of academic data. However, no prior research has employed a cluster-pattern algorithm to assess the similarities and differences between 2 clusters in networks. The study goals are 2-fold: to create a cluster-pattern comparison algorithm tailored for bibliometric analysis and to apply this algorithm in presenting clusters of countries, institutes, departments, authors (CIDA), and keywords on journal articles during and after COVID-19. We analyzed 9499 and 5943 articles from the Journal of Medicine (Baltimore) during and after COVID-19 in 2020 to 2021 and 2022 to 2023, sourced from the Web of Science (WoS) Core Collection. Follower-leading clustering algorithm (FLCA) was compared to other 8 counterparts in cluster validation and effectiveness and a cluster-pattern-comparison algorithm (CPCA) was developed using the similarity coefficient, collaborative maps, and thematic maps to evaluate CIDA cluster patterns. The similarity coefficients were categorized as identical, similar, dissimilar, or different for values above 0.7, between 0.5 and 0.7, between 0.3 and 0.5, and below 0.3, respectively. Both stages displayed similar trends in annual publications and average citations, although these trends are decreasing. The peak publication year was 2020. Similarity coefficients of cluster patterns in these 2 stages for CIDA entities and keywords were 0.73, 0.35, 0.80, 0.02, and 0.83, respectively, suggesting the existence of identical patterns (>0.70) in countries, departments, and keywords plus, but dissimilar (
ISSN:0025-7974
1536-5964
DOI:10.1097/MD.0000000000037530