Identification and causal analysis of predatory open access journals based on interpretable machine learning

Jinhong Wu, Tianye Liu, Keliang Mu () and Lei Zhou
Additional contact information
Jinhong Wu: Wuhan Textile University
Tianye Liu: Wuhan Textile University
Keliang Mu: Wuhan Textile University
Lei Zhou: Wuhan Textile University

Scientometrics, 2024, vol. 129, issue 4, No 8, 2158 pages

Abstract: Abstract Predatory journals have been a recent phenomenon, drawing attention from the academic community in the last decade. However, as the open access (OA) movement has gained momentum, the indiscriminate growth of predatory journals has had significant negative impacts on academic communication, scholarly publishing, and effective utilization of scientific resources. This rampant growth poses a serious threat to the healthy development of the OA movement and also undermines the integrity of research and the research ecosystem. Identifying predatory journals from the massive number of OA journals would assist scholars in evading negative consequences in areas of monetary investment, reputation, academic influence, and occupational advancement. Traditional methods for identifying predatory journals have relied heavily on the knowledge of domain experts. However, a large number of predatory journals exhibit latent and covert characteristics, and the growth rate of OA journals is extremely rapid, making it difficult for experts to identify these predatory journals from the vast number of OA journals. This paper proposes an interpretable machine learning model for early warning of predatory OA journals, which identifies predatory journals through the ensemble of multiple machine learning algorithms. Specifically, the proposed methodology first constructs an OA journal early warning indicator system and integrates multiple machine learning algorithms to compute the early warning values of OA journals. Then, the SHAP interpretable framework is introduced to analyze the causal factors of the early warning risks in a novel way. To verify the accuracy of the model's causal factors, we conduct a comparative analysis of domestic and foreign medical OA journals using case studies. The empirical analysis conducted in this study demonstrates the efficacy of the ensemble algorithm in accurately identifying the risk of predatory OA journals.

Keywords: Open access; Journal alerts; Causal analysis; Interpretable machine learning (search for similar items in EconPapers)
Date: 2024
References: Add references at CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s11192-024-04969-6 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:scient:v:129:y:2024:i:4:d:10.1007_s11192-024-04969-6

Ordering information: This journal article can be ordered from
http://www.springer.com/economics/journal/11192

DOI: 10.1007/s11192-024-04969-6

Access Statistics for this article

Scientometrics is currently edited by Wolfgang Glänzel

More articles in Scientometrics from Springer, Akadémiai Kiadó
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().