Archives des auteurs : Hans Dillaerts

‘As of my last knowledge update’: How is content generated by ChatGPT infiltrating scientific papers published in premier journals?

Posté le par

Author : Artur Strzelecki

The aim of this paper is to highlight the situation whereby content generated by the large language model ChatGPT is appearing in peer-reviewed papers in journals by recognized publishers. The paper demonstrates how to identify sections that indicate that a text fragment was generated, that is, entirely created, by ChatGPT. To prepare an illustrative compilation of papers that appear in journals indexed in the Web of Science and Scopus databases and possessing Impact Factor and CiteScore indicators, the SPAR4SLR method was used, which is mainly applied in systematic literature reviews.

Three main findings are presented: in highly regarded premier journals, articles appear that bear the hallmarks of the content generated by AI large language models, whose use was not declared by the authors (1); many of these identified papers are already receiving citations from other scientific works, also placed in journals found in scientific databases (2); and, most of the identified papers belong to the disciplines of medicine and computer science, but there are also articles that belong to disciplines such as environmental science, engineering, sociology, education, economics and management (3).

This paper aims to continue and add to the recently initiated discussion on the use of large language models like ChatGPT in the creation of scholarly works.

URL : ‘As of my last knowledge update’: How is content generated by ChatGPT infiltrating scientific papers published in premier journals?

DOI : https://doi.org/10.1002/leap.1650

Patent research in academic literature. Landscape and trends with a focus on patent analytics

Posté le par

Authors : Cristian Mejia, Yuya Kajikawa

Patent analytics is crucial for understanding innovation dynamics and technological trends. However, a comprehensive overview of this rapidly evolving field is lacking. This study presents a data-driven analysis of patent research, employing citation network analysis to categorize and examine research clusters. Here, we show that patent research is characterized by interconnected themes spanning fundamental patent systems, indicator development, methodological advancements, intellectual property management practices, and diverse applications.

We reveal central research areas in patent strategies, technological impact, and patent citation research while identifying emerging focuses on environmental sustainability and corporate innovation. The integration of advanced analytical techniques, including AI and machine learning, is observed across various domains. This study provides insights for researchers and practitioners, highlighting opportunities for cross-disciplinary collaboration and future research directions.

URL : Patent research in academic literature. Landscape and trends with a focus on patent analytics

DOI : https://doi.org/10.3389/frma.2024.1484685

Comparison of Feature Learning Methods for Metadata Extraction from PDF Scholarly Documents

Posté le par

Authors : Zeyd Boukhers, Cong Yang

The availability of metadata for scientific documents is pivotal in propelling scientific knowledge forward and for adhering to the FAIR principles (i.e. Findability, Accessibility, Interoperability, and Reusability) of research findings. However, the lack of sufficient metadata in published documents, particularly those from smaller and mid-sized publishers, hinders their accessibility. This issue is widespread in some disciplines, such as the German Social Sciences, where publications often employ diverse templates. To address this challenge, our study evaluates various feature learning and prediction methods, including natural language processing (NLP), computer vision (CV), and multimodal approaches, for extracting metadata from documents with high template variance.

We aim to improve the accessibility of scientific documents and facilitate their wider use. To support our comparison of these methods, we provide comprehensive experimental results, analyzing their accuracy and efficiency in extracting metadata. Additionally, we provide valuable insights into the strengths and weaknesses of various feature learning and prediction methods, which can guide future research in this field.

URL : Comparison of Feature Learning Methods for Metadata Extraction from PDF Scholarly Documents

Arxiv : https://arxiv.org/abs/2501.05082

A role for qualitative methods in researching Twitter data on a popular science article’s communication

Posté le par

Authors : Travis Noakes, Corrie Susanna Uys, Patricia Ann Harpur, Izak van Zyl

Big Data communication researchers have highlighted the need for qualitative analysis of online science conversations to better understand their meaning. However, a scholarly gap exists in exploring how qualitative methods can be applied to small data regarding micro-bloggers’ communications about science articles. While social media attention assists with article dissemination, qualitative research into the associated microblogging practices remains limited. To address these gaps, this study explores how qualitative analysis can enhance science communication studies on microblogging articles.

Calls for such qualitative approaches are supported by a practical example: an interdisciplinary team applied mixed methods to better understand the promotion of an unorthodox but popular science article on Twitter over a 2-year period. While Big Data studies typically identify patterns in microbloggers’ activities from large data sets, this study demonstrates the value of integrating qualitative analysis to deepen understanding of these interactions. In this study, a small data set was analyzed using NVivo™ by a pragmatist and MAXQDA™ by a statistician.

The pragmatist’s multimodal content analysis found that health professionals shared links to the article, with its popularity tied to its role as a communication event within a longstanding debate in the health sciences. Dissident professionals used this article to support an emergent paradigm. The analysis also uncovered practices, such as language localization, where a title was translated from English to Spanish to reach broader audiences.

A semantic network analysis confirmed that terms used by the article’s tweeters strongly aligned with its content, and the discussion was notably pro-social. Meta-inferences were then drawn by integrating the findings from the two methods. These flagged the significance of contextualizing the sharing of a health science article in relation to tweeters’ professional identities and their stances on health-related issues. In addition, meta-critiques highlighted challenges in preparing accurate tweet data and analyzing them using qualitative data analysis software. These findings highlight the valuable contributions that qualitative research can make to research involving microblogging data in science communication. Future research could critique this approach or further explore the microblogging of key articles within important scientific debates.

URL : A role for qualitative methods in researching Twitter data on a popular science article’s communication

DOI : https://doi.org/10.3389/frma.2024.1431298

How to build an Open Science Monitor based on publications? A French perspective

Posté le par

Authors : Laetitia Bracco, Eric Jeangirard, Anne L’Hôte, Laurent Romary

Many countries and institutions are striving to develop tools to monitor their open science policies. Since 2018, with the launch of its National Plan for Open Science, France has been progressively implementing a monitoring framework for its public policy, relying exclusively on reliable, open, and controlled data. Currently, this monitoring focuses on research outputs, particularly publications, as well as theses and clinical trials.

Publications serve as a basis for analyzing other dimensions, including research data, code, and software. The metadata associated with publications is therefore particularly valuable, but the methodology for leveraging it raises several challenges. Here, we briefly outline how we have used this metadata to construct the French Open Science Monitor.

URL : How to build an Open Science Monitor based on publications? A French perspective

Arxiv : https://arxiv.org/abs/2501.02856

Rejected papers in academic publishing: Turning negatives into positives to maximize paper acceptance

Posté le par

Authors : Jaime A. Teixeira da Silva, Maryna Nazarovets

There are ample reasons why papers might get rejected by peer-reviewed journals, and the experience can be, especially for those who have had little experience, sobering. When papers get rejected a number of times, that may signal that there are problems with the paper (e.g., weak methodology or lack of robust analyses), that it is insufficiently developed, is poorly written, or that it is too topic-specific and needs to find an appropriate niche journal.

In the case of a single or multiple rejections, whenever there is feedback from a journal, as well as reasons for rejection, this provides a useful signal for improving the paper before it is resubmitted to another journal. This article examines literature related to the rejection of papers in academic journals, encompassing the opinions and experiences offered by authors, as well as advice suggested by editors, allowing readers and authors who experience rejections to reflect on the possible reasons that may have led to that outcome.

Many papers related to this topic were published as editorials or opinions, offering advice on how to improve aspects of a submitted paper in order to increase its chances of acceptance.

URL : Rejected papers in academic publishing: Turning negatives into positives to maximize paper acceptance

DOI : https://doi.org/10.1002/leap.1649

 

Recycling Research Without (Self-)Plagiarism: The Importance of Context and the Case of Conference Contributions

Posté le par

Authors : Gert Helgesson, Jonas Åkerman, Sara Belfrage

In this paper, we clarify the notions of plagiarism and self-plagiarism and show that a rather straightforward observation about these notions has important implications for the admissibility of recycling research outputs. The key point is that contextual variation must be taken into account in normative assessments of recycling research outputs, and we illustrate this with some examples.

In particular, we apply the analysis in order to dissolve a disagreement about the proper handling of submissions to conferences. Some researchers are comfortable with sending the same contribution to several conferences, while others find that unacceptable and a clear deviation from good research practise. We take a closer look at the arguments regarding whether it is acceptable or not to make the same conference contribution more than once, including the argument that submitting the same contribution more than once would amount to self-plagiarism.

We argue that contextual variation must be taken into account, in accordance with our previous analysis, and conclude that whether or not a duplication of a conference contribution deviates from good research practise depends on what significance is ascribed to it in the specific case. We conclude with some practical recommendations, emphasising for example, the importance of being explicit and clear on this point, and encourage conference organisers to provide opportunities to specify relevant facts in the submission.

URL : Recycling Research Without (Self-)Plagiarism: The Importance of Context and the Case of Conference Contributions

DOI : https://doi.org/10.1002/leap.1653