Specialist magazines with a high proportion of suspicious articles, characterized by science integrity startups

Specialist magazines with a high proportion of suspicious articles, characterized by science integrity startups

Which scientific publishers and magazines are most affected by fraudulent or questionable research - and which have been the least done to clean up their portfolios? A technology start-up that was founded to help publishers identify potentially problematic articles has some answers and shared its first findings with nature.

The Knowledge Ingritrity website Argos, which was started in September by Scitility, a technology company based in Sparks, Nevada, publishes risk reviews for articles based on the authors' publication data and whether the article has a strong references to already withdrawn research. An article classified as "highly risky" can have several authors whose other studies have been withdrawn due to misconduct. A high number of points does not prove that an article is of low quality, but indicates that a review could be useful.

Argos is part of a growing number of instruments for the integrity check in research that search for warning signals in articles. These include the PaperMill Alarm , developed by Clear Skies, and Signals from Research Signals, both from London. Since the developers of such software sell their manuscript screening tools to publishers, they are generally reserved when it comes to the mention of affected magazines. Argos, which offers individuals free accounts and comprehensive access for integrity auditors and journalists, is the first to provide public insights.

"We wanted to develop a technology that can recognize hidden patterns and create transparency for the industry," says Erik de Boer, co -founder of Scitility, which is based in Roosendaal, Netherlands.

risk assessment of publishers

The analysis of Argos shows that the publisher Hindawi - a now closed subsidiary of the London publisher Wiley - has the highest number and the highest proportion of already withdrawn articles (see "Publisher in Risk"). That is not surprising, because Wiley has In the past two years over 10,000 articles published by Hindawi , in response to concerns that have been expressed by editors and examiners; This corresponds to more than 4% of the brand's overall portfolio in the past decade. One of his magazines, the Evidence-Based Complementary and Alternative Medicine, has withdrawn 741 articles, which makes up more than 7% of its production.

The risk assessments of Argos mark more than a thousand remaining Hindawi articles-another 0.65%-as still "highly risky". This indicates that, although Wiley has done a lot to clean up his portfolio, the problem may not yet have been completely solved. The publisher informed Nature that he greeted Argos and similar tools and is working on fixing the problems with Hindawi.

Andere Verlage scheinen viel mehr Untersuchungen anstellen zu müssen, da die Anzahl der Rückzüge im Verhältnis zur Anzahl der von Argos gekennzeichneten hochriskanten Artikel gering ist (Verlage könnten einige dieser Artikel bereits untersucht und entschieden haben, dass keine Maßnahmen erforderlich sind).

The publisher Elsevier based in Amsterdam has around 5,000 withdrawals, but over 11,400 high -risk articles, according to the analysis of Argos by nature - although all of these together only make up a little more than 0.2% of the publisher's production in the past decade. The publisher MDPI has withdrawn 311 items, but has more than 3,000 high -risk articles - about 0.24% of its production. Springer Nature has more than 6,000 withdrawals and more than 6,000 high -risk articles; about 0.3% of its production. (The Nature news team is independent of its publisher.)

In answers to inquiries, all publishers, which are characterized as the largest providers of high -risk articles, are used to use research integrity, use technologies to check submitted articles and that their retreats to clean up problematic content.

Springer Nature reported that it introduced two tools in June that have helped to recognize hundreds of fake manuscripts since then; Several publishers raised their collaboration in a Common integrity center , which offers software that can characterize suspicious articles. Jisuk Kang, a publishing manager at MDPI in Basel, Switzerland, says that products such as Argos provide broad references to potential problems, but notes that the publisher cannot check the accuracy or reliability of the numbers on the website. It adds that the largest publishers and magazines would inevitably have a higher number of high -risk articles, so that the proportion of production is a better key figure.

The publishing stamps with the highest proportions of high-risk articles in their portfolios are, according to Argos data Impact Journals (0.82%), spoard (0.77%) and IvYspring (0.67%). Impact Journals reports that, although his magazines in the past have had problems, they have now improved their integrity. The publisher states that in the magazine Oncotarget in the past two years " 0% irregularities ", which is due to the introduction of image examination instruments such as Image Twin, which have only been available in recent years. Portland Press, which has 0.41% high -risk articles in his portfolio, states that he had taken correct measures to improve the strict exams.

Risk reviews of magazines

Argos also offers numbers for individual magazines. It is not surprising that Hindawi titles stand out both with regard to the number and the proportion of the withdrawn work, while other magazines have many of the work identified by Argos (see "Magazines in Risk"). According to volume, the Mega magazine Scientific Reports by Springer Nature with 450 high-risk articles and 231 withdrawals, which together is about 0.3% of its production. On October 16, a group of examiners wrote a Open letter to Springer Nature , in which concerns about problematic articles were expressed in the magazine.

In the answer, Chris Graf, head of research integrity at Springer Nature, says that the magazine examines any problem mentioned. He adds that the proportion of content that has been highlighted is relatively low compared to their size.

magazines with particularly large discrepancies between the number of withdrawn work and the potentially suspicious articles include the sustainability magazine of MDPI (20 withdrawals and 312 high -risk articles; 0.4% of production) and the Materials Today Proceedings (28 withdrawals and 308 high -risk articles; 0.8% of production). Elseviers Biomedicine & Pharmacotherapy has the highest proportion of high -risk articles - 1.61% of its production.

"The volume of fraudulent materials increases significantly, supported by systematic manipulations such as 'Papermills', which produce fraudulent content for commercial purposes, and AI generated content," says a spokesman for Elsevier and adds that in response "we increase our investments in human supervision, specialist knowledge and technology.

open data

The developers of Argos emphasize that the website is based on open data collected by others. The sources include the Retraction Watch website, which leads a database with withdrawn articles - free of charge via an agreement with the non -profit organization CrossRef - which contains the reasons for a retreat, so that tools that check the author data can concentrate on retreats. The analysis is also based on Records of articles that have strongly referred to withdrawn work , which was put together by Guillaume Cabanac, an computer scientist at the University of Toulouse in France.

Although Argos also follows analysts, the Networks of authors with a history of misconduct Focus on other integrity testing tools also characterize articles based on suspicious content, such as tight textual similarities to fake work or "disturbed phrases", a term characterized by Cabanac when authors Make a strange choice of words to avoid that plagiarism identification systems are activated.

"Both approaches have their justification, but the identification of researchers of researchers who participate in misconduct could be more valuable," says James Butcher, a former publisher of Nature magazines and the Lancet, who now leads the advisory company Journalology in Liverpool, UK. "This is because AI supported writing tools could be used to support fraudsters to avoid obvious textual information," he adds. Butcher adds that many large publishers have developed or acquired their own integrity tools to check various warning signals in manuscripts.

One of the trickiest problems for integrity tools that are mainly based on the retreats of authors is the correct distinction between authors with a similar name - a problem that argos could distort numbers. "The problem of author discrimination is the biggest problem that the industry has," says Adam Day, founder of Clear Skies.

de Boer, who previously worked at Springer Nature, says that everyone can create an account to use Argos for free, but plans to sell Scitility to a version of the tool to large publishers and institutions that could integrate it directly into their manuscript screening workflows.

Butcher praises the transparency of the Argos team. "There must be more visibility for magazines and publishers that take abbreviations and do not exercise adequate care in the published and monetized work," he says.