Authorship Attribution Research Articles

Semantic Clustering and Transfer Learning in Social Media Texts Authorship Attribution

Semantic Clustering and Transfer Learning in Social Media Texts Authorship Attribution

AGI-P: A Gender Identification Framework for Authorship Analysis Using Customized Fine-Tuning of Multilingual Language Model

AGI-P: A Gender Identification Framework for Authorship Analysis Using Customized Fine-Tuning of Multilingual Language Model

Nanomedicine-based cancer immunotherapy: a bibliometric analysis of research progress and prospects.

Despite the increasing number of studies on nanomedicine-based cancer immunotherapy, the overall research trends in this field remain inadequately characterized. This study aims to evaluate the research trends and hotspots in nanomedicine-based cancer immunotherapy through a bibliometric analysis. As of March 31, 2024, relevant publications were retrieved from the Web of Science Core Collection. Analytical tools including VOSviewer, CiteSpace, and an online bibliometric analysis platform were employed. A total of 5,180 publications were analyzed. The study reveals geographical disparities in research output, with China and the United States being the leading contributors. Institutionally, the Chinese Academy of Sciences, University of Chinese Academy of Sciences, and Sichuan University are prominent contributors. Authorship analysis identifies key researchers, with Liu Zhuang being the most prolific author. "ACS Nano" and the "Journal of Controlled Release and Biomaterials" are identified as the leading journals in the field. Frequently occurring keywords include "cancer immunotherapy" and "drug delivery." Emerging frontiers in the field, such as "mRNA vaccine," "sonodynamic therapy," "oral squamous cell carcinoma," "STING pathway,"and "cGAS-STING pathway," are experiencing rapid growth. This study aims to provide new insights to advance scientific research and clinical applications in nanomedicine-based cancer immunotherapy.

A Comparative text similarity analysis of the works of Bernard Mandeville

Text similarity analysis entails studying identical and closely similar text passages across large corpora, with a particular focus on intentional and unintentional borrowing patterns. At a larger scale, detecting repeated passages takes on added importance, as the same text can convey different meanings in different contexts. This approach offers numerous benefits, enhancing intellectual and literary scholarship by simplifying the identification of textual overlaps. Consequently, scholars can focus on the theoretical aspects of reception with an expanded corpus of evidence at their disposal. This article adds to the expanding field of historical text reuse, applying it to intellectual history and showcasing its utility in examining reception, influence, popularity, authorship attribution, and the development of tools for critical editions. Focused on the works and various editions of Bernard Mandeville (1670–1733), the research applies comparative text similarity analysis to explore his borrowing habits and the reception of his works. Systematically examining text reuses across several editions of Mandeville’s works, it provides insights into the evolution of his output and influences over time. The article adopts a forward-looking perspective in historical research, advocating for the integration of archival and statistical evidence. This is illustrated through a detailed examination of the attribution of Publick Stews to Mandeville. Analysing cumulative negative evidence of borrowing patterns suggests that Mandeville might not have been the author of the piece. However, the article aims not to conclude the debate but rather to open it up, underscoring the importance of taking such evidence into consideration. Additionally, it encourages scholars to incorporate text reuse evidence when exploring other cases in early modern scholarship. This highlights the adaptability and scalability of text similarity analysis as a valuable tool for advancing literary studies and intellectual history.

I repeat therefore I am: The parasyntactic perspective

Abstract The text presents a theoretical platform and a case study of a new method for authorship attribution based on an author’s specific low-frequency lexicon. It will be shown that an author’s text is largely context-independent and is constructed by the author’s habit based on the regular repetition of certain topics or modes of expression. The author’s idiosyncratic way of choosing between synonymous linguistic devices in the text happens at a distance of several word forms or sentence units. This means that texts themselves are constructed using a much wider range of repetitions than expected and that the structure of the text above the level of intersentential linking is determined by a specific group of words (functional but above all content words) obligatorily used by the author in the formulation of the text. The newly introduced method can be used to attribute authorship by relying on the specific linguistic imprint of the author in the text (in this context, we talk about parasyntactic linguistic level). The method is compared with a function-word-based method.

Transformer-Based Composite Language Models for Text Evaluation and Classification

Parallel natural language processing systems were previously successfully tested on the tasks of part-of-speech tagging and authorship attribution through mini-language modeling, for which they achieved significantly better results than independent methods in the cases of seven European languages. The aim of this paper is to present the advantages of using composite language models in the processing and evaluation of texts written in arbitrary highly inflective and morphology-rich natural language, particularly Serbian. A perplexity-based dataset, the main asset for the methodology assessment, was created using a series of generative pre-trained transformers trained on different representations of the Serbian language corpus and a set of sentences classified into three groups (expert translations, corrupted translations, and machine translations). The paper describes a comparative analysis of calculated perplexities in order to measure the classification capability of different models on two binary classification tasks. In the course of the experiment, we tested three standalone language models (baseline) and two composite language models (which are based on perplexities outputted by all three standalone models). The presented results single out a complex stacked classifier using a multitude of features extracted from perplexity vectors as the optimal architecture of composite language models for both tasks.

Epistemic Modality Constructions as Stable Idiolectal Features: A Cross-genre Study of Spanish

Forensic authorship analysis is based on two assumptions: that every individual has a unique idiolect, and that features characteristic of that idiolect will recur with a relatively stable frequency. Yet, a speaker’s language can change with age, affective states, according to audience, or genre. Thus, studies on authorship analysis should adopt the theory that while some linguistic parameters of an idiolect can remain stable, others can change depending on various circumstances. This investigation, which takes a constructional and functional-based approach to discourse-level phenomena in idiolectal stability, analyzes cross-genre data produced by nine Mexican participants throughout a twelve-year time span. This study contributes to a greater understanding of the linguistic elements that survive genre effects and are potentially useful in both investigative and evidential forensic linguistic work. We provide a detailed description of linguistic features, their specific values, and context-dependent interpretation, keeping in mind the context of expert linguistic testimony, with its preference for methods which “employ linguistically motivated analyses in combination with quantitative tools” (Solan & Tiersma, 2004, p.463).Our findings show that idiolectal style tends to remain stable across genres and communication modes in epistemic modality constructions. Epistemic markers —specifically, markers indicating low commitment by the speaker (e.g., no sé ‘I don’t know’) or expressing indirectness when introducing the illocutionary force (e.g., la verdad [es que] ‘the truth [is that]’)— display idiolectal stability, as these markers seem to be the most effective in terms of allowing speakers to strategically manifest the extent of their knowledge regarding what is said.

Who could be behind QAnon? Authorship attribution with supervised machine-learning

Abstract A series of social media posts on 4chan then 8chan, signed under the pseudonym ‘Q’, started a movement known as QAnon, which led some of its most radical supporters to violent and illegal actions. To identify the person(s) behind Q, we evaluate the coincidence between the linguistic properties of the texts written by Q and to those written by a list of suspects provided by journalistic investigation. To identify the authors of these posts, serious challenges have to be addressed. The ‘Q drops’ are very short texts, written in a way that constitute a sort of literary genre in itself, with very peculiar features of style. These texts might have been written by different authors, whose other writings are often hard to find. After an online ethnography of the movement, necessary to collect enough material written by these thirteen potential authors, we use supervised machine learning to build stylistic profiles for each of them. We then performed a ‘rolling analysis’, looking repeatedly through a moving window for parts of Q’s writings matching our profiles. We conclude that two different individuals, Paul F. and Ron W., are the closest match to Q’s linguistic signature, and they could have successively written Q’s texts. These potential authors are not high-ranked personality from the US administration, but rather social media activists.

Malware authorship attribution: Unmasking the culprits behind malicious software

With the digital age ushering in an unprecedented proliferation of malware, accurately attributing these malicious software variants to their original authors or affiliated groups has emerged as a crucial endeavor in cybersecurity. This study delves into the intricacies of malware authorship attribution by combining traditional analytical techniques with advanced machine learning methodologies. An integrated approach, encompassing static and dynamic analyses, yielded promising results in the challenging realm of malware attribution. Despite the encouraging outcomes, the research highlighted the multifaceted complexities involved, especially considering the sophisticated obfuscation techniques frequently employed by attackers. This paper emphasizes the merits of a holistic attribution model and underscores the importance of continuous innovation in the face of an ever-evolving threat landscape.

Authorship Attribution on Short Texts in the Slovenian Language

The study investigates the task of authorship attribution on short texts in Slovenian using the BERT language model. Authorship attribution is the task of attributing a written text to its author, frequently using stylometry or computational techniques. We create five custom datasets for different numbers of included text authors and fine-tune two BERT models, SloBERTa and BERT Multilingual (mBERT), to evaluate their performance in closed-class and open-class problems with varying numbers of authors. Our models achieved an F1 score of approximately 0.95 when using the dataset with the comments of the top five users by the number of written comments. Training on datasets that include comments written by an increasing number of people results in models with a gradually decreasing F1 score. Including out-of-class comments in the evaluation decreases the F1 score by approximately 0.05. The study demonstrates the feasibility of using BERT models for authorship attribution in short texts in the Slovenian language.

Women in the Lab, Men in the Field? Correlations between Gender and Research Topics at Three Major Archaeology Conferences

ABSTRACT Rising interest in gender equality in society has resulted in greater scrutiny of gender inequality in academic communities. Analysis of authorship of peer-reviewed publications shows that archaeology, like other academic fields, has long been dominated by men. We ask if gender disproportionality is evident in the topics presented by archaeologists at major conferences, particularly the Society of American Archaeology (SAA), the European Association of Archaeologists (EAA), and the Computer Applications and Quantitative Methods in Archaeology (CAA) meetings. Does participants’ gender correlate with the topics of their presentations? We analyzed presenters’ names in published programs to infer gender. We used machine learning to identify topics from presentation titles. We found distinctive topics that are strongly associated with women, such as cultural heritage, GIS, and isotope analyses. Awareness of these correlations between research topics and gender is important to ensure equitable participation in archaeology and unbiased access to training opportunities for students.

Attributions of Authorship in the Gentleman’s Magazine, 1831–1835: A Supplement to the Union List

Attributions of Authorship in the <i>Gentleman’s Magazine</i>, 1831–1835: A Supplement to the <i>Union List</i>

Are we teaching the health impacts of climate change in a clinically relevant way? A systematic narrative review of biomechanism-focused climate change learning outcomes in medical curricula

Purpose Introducing biomedical approaches to the health impacts of climate change can improve medical student engagement with relevant climate-related issues, improve the development of medical schemas, and minimise displacement into crowded medical curricula. This paper aims to systematically review the medical education curricula related to climate change, with a particular focus on systems-based biomechanisms for the health impacts of climate change. We do this to provide a clear agenda for further development of learning outcomes (LOs) in this area to maximize the clinical applicability of this knowledge. Material and Methods A systematic review was undertaken following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA; Liberati et al. 2009) guidelines for both the published and grey literature. Five databases (PubMed, SCOPUS, ERIC, Open Access Thesis and Dissertation, and Proquest Global Dissertation and Theses) were searched for works published between 2011 and June 2023. Full texts that contained LOs were the main inclusion criteria for the final review. Descriptive and content extraction guided the final narrative synthesis. Results Analysis indicated that biomechanism-related LOs represented about 25% of each published LO set, on average. These outcomes were primarily at the “understand” level of Bloom’s taxonomy and were spread across a range of body systems and climate-change aspects. Infectious diseases and extreme heat were strong focuses. Authorship analysis indicated that the majority of these sets of published LOs are from Western contexts and authored by researchers and educators with medical and population health qualifications. Conclusions Biomechanism-focused teaching about the health impacts of climate change is relatively rare in published curricula. Of the available sets of LOs, the majority are sourced from Western authors and are focused on a fairly circumscribed set of biomedical topics. There is scope to both broaden and deepen curriculum in this area, and we would recommend the field prioritise collaboration with medical educators from the Global South, where the effects of climate change are already the most acutely felt.

Same or Different? Diff-Vectors for Authorship Analysis

In this article, we investigate the effects on authorship identification tasks (including authorship verification, closed-set authorship attribution, and closed-set and open-set same-author verification) of a fundamental shift in how to conceive the vectorial representations of documents that are given as input to a supervised learner. In “classic” authorship analysis, a feature vector represents a document, the value of a feature represents (an increasing function of) the relative frequency of the feature in the document, and the class label represents the author of the document. We instead investigate the situation in which a feature vector represents an unorderedpairof documents, the value of a feature represents the absolute difference in the relative frequencies (or increasing functions thereof) of the feature in the two documents, and the class label indicates whether the two documents are from the same author or not. This latter (learner-independent) type of representation has been occasionally used before, but has never been studied systematically. We argue that it is advantageous, and that, in some cases (e.g., authorship verification), it provides a much larger quantity of information to the training process than the standard representation. The experiments that we carry out on several publicly available datasets (among which one that we here make available for the first time) show that feature vectors representing pairs of documents (that we here callDiff-Vectors) bring about systematic improvements in the effectiveness of authorship identification tasks, and especially so when training data are scarce (as it is often the case in real-life authorship identification scenarios). Our experiments tackle same-author verification, authorship verification, and closed-set authorship attribution; while DVs are naturally geared for solving the 1st, we also provide two novel methods for solving the 2nd and 3rd that use a solver for the 1st as a building block. The code to reproduce our experiments is open-source and available online.1

Identifying critical gaps in research to advance global surgery by 2030: a systematic mapping review

Progress on surgical system strengthening has been slow due to a disconnect between evidence generation and the information required for effective policymaking. This systematic mapping review sought to assess critical research gaps in the field of global surgery guided by the World Health Organisation Health Systems building block framework, analysis of authorship and funding patterns, and an exploration of emerging research partnership networks. Literature was systematically mapped to identify, screen, and synthesize results of publications in the global surgery field between 2015 and March 2022. We searched four databases and included literature published in seven languages. A social network analysis determined the network attributes of research institutions and their transient relationships in shaping the global surgery research agenda. We identified 2,298 relevant studies out of 92,720 unique articles searched. Research output increased from 453 in 2015-16 to 552 in 2021-22, largely due to literature on Covid-19 impacts on surgery. Sub-Saharan Africa (792/2298) and South Asia (331/2298) were the most studied regions, although high-income countries represented a disproportionate number of first (42%) and last (43%) authors. Service delivery received the most attention, including the surgical burden and quality and safety of services, followed by capacity-building efforts in low- and middle-income countries. Critical research in economics and financing, essential infrastructure and supplies, and surgical leadership necessary to guide policy decisions at the country level were lacking. Global surgical systems remain largely under-researched. Knowledge diffusion requires an emphasis on developing sustainable research partnerships and capacity across low- and middle-income countries. A renewed focus must be given to equipping countries with tools for effective decision-making to enhance investments in high-quality surgical services.

A bibliometric study and visualization analysis of ferroptosis-inducing cancer therapy

Ferroptosis is a form of regulated cell death that was first formally proposed a decade ago. While its role in cancer cell death was initially understudied, it has recently gained considerable interest from researchers. In recent years, a growing number of studies have focused on the role of ferroptosis in cancer progression, with the goal of developing novel ferroptosis-inducing cancer therapies. This study aims to present the developmental trend and hotspots of research on ferroptosis-inducing cancer therapy using bibliometric analysis. A literature search was conducted using the Web of Science Core Collection on October 1st, 2022, to retrieve articles and reviews pertaining to ferroptosis and cancer published from 2012 to 2022. Microsoft Excel 2016, VOSviewer 1.6.18 and CiteSpace (version 6.1. R6) were utilized to conduct the bibliometric analysis of publication trends, authorship, and citation networks, with a focus on identifying countries, institutions, journals, and authors contributing to the field. These analyses were used to predict future trends in this area. A total of 2839 articles were identified and extracted for analysis. The number of publications has increased almost every year, with a sharp increase after 2018. China produced the most publications in this area, followed by the United States. Central South University was the institution that published the most papers. Frontiers in Oncology was the journal with the highest number of publications, while Cell had the greatest impact factor. Daolin Tang was the most productive author and Dixon SJ was the most influential author. Co-occurrence and burst analyses of keywords and references were conducted to identify the developmental trends and hotspots in ferroptosis-inducing cancer therapy research. Main research directions have shifted from investigating the mechanism of ferroptosis to developing novel ferroptosis-targeting cancer therapies. Emerging topicsfocus on the role of ferroptosis in solid tumor therapy. Based on our bibliometric analysis, we predict that research on ferroptosis in cancer therapy will continue to be a hot topic in the future, with a growing number of treatment modalities related to ferroptosis being developed. Our study provides valuable insights into the current state and future trends of research in this field, serving as a useful guide for researchers seeking to make important contributions in this area.

Reproducibility and non-traditional authorship attribution: Invitatio ad Arma

Reproducibility and non-traditional authorship attribution: Invitatio ad Arma

THE ABUSE OF AUTHORSHIP IN THE BIOMEDICAL LITERATURE

The ‘publish or perish’ culture has induced an unhealthy aspect of academic publishing in the biomedical sciences, namely of illegitimate authorship. Such a position can be obtained when an authorship slot is offered as a gift (e.g., to a sponsor, or researcher in a senior position), without their intellectual or scientific participation, or it can be hidden (ghost authorship), in which the paper (or parts thereof) is written by a third party (individual, or company). In a more industrialized setting, ghost authorship takes place via the sale of papers using “paper mills”, including of specific author slots (i.e., positions in a line of authors). While author-based persistent identifiers like ORCID, or authorship attribution schemes like CRediT, sound noble and offer some form of validation, those systems still operate on a culture of blind trust (in submitting authors). This paper debates a few of the authorship-related issues currently plaguing biomedical journals.

‘A shifting precipice of unsettled law’?

This paper provides an up-to-date summary of how the US legal system treats evidence that uses the forensic stylistics method of authorship attribution analysis. Many scholars in both law and linguistics have written about some of the more notable cases on this issue, but none have attempted to trace the entire line of relevant case law since the advent of modern forensic stylistics, and relatively few summaries of any case law have appeared in the last ten years. It is hoped that a fresh look with updated legal research can add new insights for litigants, lawyers and linguists alike.

Authorship attribution of comments in Portuguese extracted from Reddit

Authorship attribution of comments in Portuguese extracted from Reddit