Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
144 tokens/sec
GPT-4o
8 tokens/sec
Gemini 2.5 Pro Pro
46 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Detecting a network of hijacked journals by its archive (2101.01224v2)

Published 4 Jan 2021 in cs.DL and cs.SI

Abstract: This study describes a method to detect hijacked journals based on the analysis of the archives of clone journals. This approach is most effective in discovering a network of hijacked journals that have the same organizer(s). Analysis of the archives of clone journals allowed to detect 62 URLs of hijacked journals. It also provided the possibility to predict two clone websites before they became operational. This study shows that most detected hijacked journals represent a network of clone journals organized by one or several fraudulent individuals. The information and content of nine legitimate journals were compromised in international and national scientometric databases.

Citations (30)

Summary

  • The paper introduces an archive-based method to detect hijacked journal networks, successfully identifying 62 journals linked by recycled content and predicting future clones.
  • The study finds hijacked journals form organized networks identified by recycled content, suggesting its use in prediction and revealing vulnerabilities in academic databases.
  • The study calls for improved detection methods, greater institutional vigilance, and using real-time content analysis to combat organized hijacked journal networks.

Analysis of Hijacked Journals Through Archive Investigation

The paper outlines a novel approach for the detection of hijacked journals, which typically become evident only after their clone websites have been operational for a certain period. The work probes into the archives of clone journals as a primary method to uncover a network of hijacked journals managed by the same entity or group. This methodology demonstrates significant efficacy, identifying 62 URLs for such journals and even predicting two clone websites before their activation.

Methodology

The paper employs a meticulous examination of the archives of existing clone journals. It begins with manually detecting potential clones and then uses scripts to analyze titles and authors through a Google Custom Search API. This automated and manual combination enhances the identification process, allowing the detection of recurrent fraudulent practices. The approach capitalizes on the premise that the same content circulates among hijacked journals, potentially acting as a predictor of further fraudulent activities.

Results

The paper’s findings highlight that most detected hijacked journals belong to a single network, managed by either an individual or a group. Among the 62 identified URLs, 57 were distinct hijacked journals that shared substantial content with their counterparts. This content recycling reflects the fraudulent intent of these entities to minimize costs while presenting a facade of legitimacy through a consistent, albeit fake, archive.

Notably, the research identifies links between hijacked and predatory journals, underscoring the shared practice of publishing identical texts without proper authorship verification. An evident pattern emerges as some journals provide links to prestigious scientometric databases like Scopus, thereby reinforcing their deceptive credibility.

Discussion

The paper sheds light on the systematic nature of journal hijacking, indicating an organized effort akin to paper mills. These networks exploit the demand for publication among scholars facing pressures such as academic promotions and grant requirements. This business model leverages low visibility and the brief operational lifespan of clone journals to propagate compromised scientific content.

Critical observations include the prevalence of hijacked journals targeting specific niches—literature, archaeology, etc.—and journals with defunct titles, notably increasing their chances of evading immediate detection. Moreover, much of the compromised content from these clone journals was later included in legitimate scientometric databases, thus hinting at the systemic vulnerabilities these databases face.

Implications

Practically, this research underscores a need for enhanced vigilance and more sophisticated detection mechanisms in combating journal hijacking. It also calls for greater scrutiny from academic institutions and publishers, akin to the vigilance exercised in the identification of paper mills.

Theoretically, this paper expands the discourse on scientific publication ethics, proposing a new dimension in the economic theory of crime—specifically targeting digital fraud in the scholarly domain. The revelation of recurrent textual recycling suggests a potential metric for predicting and identifying fraudulent journals.

Future Directions

Future research in this domain could focus on developing real-time detection algorithms that rely on content analysis rather than circumstantial network detection. Moreover, exploring the awareness and motivations of researchers who publish in such journals could illuminate broader systemic issues in the academic publication landscape. Overall, addressing this niche of fraudulent publishing is crucial to maintaining the integrity of scientific research and publication standards globally.

X Twitter Logo Streamline Icon: https://streamlinehq.com