Search

Showing total 2 results
2 results

1. Subject-Related Research Metrics in Different Scientometrics Platforms.

Author

Khorasani, Mahyar

Subjects
*DATABASES, *COMPUTER software, *MEDICAL information storage & retrieval systems, *ACCOUNTING, *BIBLIOMETRICS, *SERIAL publications, *QUANTITATIVE research, *PSYCHOLOGY, *CITATION analysis, *MATHEMATICS, *ECONOMICS, *HUMANITIES, *SCIENCE, *AUTHORSHIP, *ALGORITHMS
Abstract

This article investigates subject-related research metrics from the most popular scientometrics platforms with the capability of citation count. Quantitative analyses have been carried out to determine the most- to the least-cited subject areas to indicate the most active and dynamic fields in research. Subject areas are classified based on statistical analyses and the number of citations received in the period of search. Numerous articles were analyzed from top-ranked journals in all research fields covered by SCImago journal classification. A comprehensive data set was generated in 27 research subject areas, including 313 subject categories and 81 high-ranked journals. Results showed that the highest number of citations were related to the sciences and science-related subject areas. The subject areas of computer science and mathematics had the lowest research metrics and the smallest number of highly cited publications. Web of Science performed better than Scopus for subject-related evaluations. Analysis of frequency data showed that Google Scholar cited fewer papers for low-cited publications than did Scopus or Web of Science. For highly cited publications, however, Google Scholar was found to have better performance. Web of Science showed the best consistency in citation coverage in most of the investigated subject areas. [ABSTRACT FROM AUTHOR]

Published
2022
Full Text
View/download PDF

2. Computational Topic Models of the Library Quarterly.

Author

Hennesy, Cody and Naughton, David

Subjects
*COMPUTER simulation, *RESEARCH, *DATA curation, *ENGLISH language, *ACADEMIC libraries, *METADATA, *LEADERSHIP, *WOMEN, *CONCEPTUAL structures, *ELECTRONIC publications, *CLUSTER analysis (Statistics), *LIBRARIANS, *ALGORITHMS, *AUTHORSHIP
Abstract

This case study demonstrates the application of an unsupervised topic modeling algorithm to 7,773 English-language articles published in the Library Quarterly from 1931 to 2015. The analysis of 85 years of the journal's output follows an exploratory data analysis framework to generate novel hypotheses about the history of LIS using topic modeling, a method for identifying clusters of co-occurring words within large collections of text. The paper closely examines two topics that suggest differences in gender representation in the journal to propose and support a new hypothesis regarding the historical inclusion of gendered objects of study in LIS literature. [ABSTRACT FROM AUTHOR]

Published
2022
Full Text
View/download PDF