Measuring the impact of novelty, bibliometric, and academic-network factors on citation count using a neural network

Xinyuan Zhang, Qing Xie, Min Song

Research output: Contribution to journalArticlepeer-review

Abstract

The factors influencing academic citations have been extensively discussed in the literature. However, few studies have investigated whether atypical recombinations of references, topics, and keywords, and academic-network factors (e.g., author citation network-related factors, co-author network-related factors, and institution citation network-related factors) are correlated with paper citation counts. Also, most previous studies have only focused on one discipline. Twenty-four factors were classified into three main categories, including the novelty of the paper, bibliometric indicators, and the influence of the academic network of authors and institutions, which have not yet been simultaneously considered. To fill this gap in the literature, a neural network model was constructed to measure the influence of these 24 factors on citation counts using the weight product of connecting neurons. The results demonstrated that the influence of novelty, bibliometric, and academic-network-related factors on citation counts vary significantly among the four studied disciplines (library & information science, nuclear science & technology, computer science & software engineering, and history). It was found that the influence of multiple factors in the novelty category on citation counts is higher than the bibliometric and academic-network categories, while the individual factor in the novelty category is not always the most influential factor (median z-score, recombination topic pairs, and recombination keyword pairs).

Original languageEnglish
Article number101140
JournalJournal of Informetrics
Volume15
Issue number2
DOIs
Publication statusPublished - 2021 May

Bibliographical note

Funding Information:
This work was supported by the National Research Foundation of Korea (NRF) and funded by the Korean Government (grant no. NRF-2019R1A2C2002577 ) and the China Scholarship Council .

Publisher Copyright:
© 2021 Published by Elsevier Ltd.

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Library and Information Sciences

Fingerprint Dive into the research topics of 'Measuring the impact of novelty, bibliometric, and academic-network factors on citation count using a neural network'. Together they form a unique fingerprint.

Cite this