Amidst the rising capabilities of generative AI to mimic specific human styles, this study investigates the ability of state-of-the-art large language models (LLMs), including GPT-4o, Gemini 1.5 Pro, and Claude Sonnet 3.5, to emulate the authorial signatures of prominent literary and political figures: Walt Whitman, William Wordsworth, Donald Trump, and Barack Obama. Utilizing a zero-shot prompting framework with strict thematic alignment, we generated synthetic corpora evaluated through a complementary framework combining transformer-based classification (BERT) and interpretable machine learning (XGBoost). Our methodology integrates Linguistic Inquiry and Word Count (LIWC) markers, perplexity, and readability indices to assess the divergence between AI-generated and human-authored text. Results demonstrate that AI-generated mimicry remains highly detectable, with XGBoost models trained on a restricted set of eight stylometric features achieving accuracy comparable to high-dimensional neural classifiers. Feature importance analyses identify perplexity as the primary discriminative metric, revealing a significant divergence in the stochastic regularity of AI outputs compared to the higher variability of human writing. While LLMs exhibit distributional convergence with human authors on low-dimensional heuristic features, such as syntactic complexity and readability, they do not yet fully replicate the nuanced affective density and stylistic variance inherent in the human-authored corpus. By isolating the specific statistical gaps in current generative mimicry, this study provides a comprehensive benchmark for LLM stylistic behavior and offers critical insights for authorship attribution in the digital humanities and social media.
The emoticon picture clarity in online service encounters has been overlooked in consumer research. Our study intends to investigate how emoticon picture clarity influences consumer service satisfaction. Across four experiments and a single-paper meta-analysis, we demonstrate that when service providers use clear rather blurred emoticon pictures to communicate with consumers, consumers will have higher service satisfaction (Study 1). This effect is attributed to the higher processing fluency induced by clear emoticon pictures, which in turn triggers greater satisfaction (Studies 2 and 3). Furthermore, this effect is weakened when consumers experience cognitive load (Study 4). These findings provide novel insights into consumers' biased evaluations of service providers and offer valuable guidance for marketers to enhance online shopping services through the strategic use of emoticon pictures.
History of scholarship and learning. The humanities, Social Sciences
Mauro Marino-Jiménez, Norma Sánchez-Chávez, Yenny Rivero-Fortón
et al.
Student performance, disciplinary innovation and teaching methodology occupy the main concerns of educational research. Therefore, there is a greater interest in gamification strategies, where digital tools facilitate the development of competitive activities and strengthening of learning. One example of this idea is the use of video games created for non-educational purposes, where disciplinary strategies and/or social skills can be developed. In this paper, the game Among Us is used to develop an educational experience at higher education. Its use helps to develop a methodology for the identification and analysis of fallacies, according to their recurrency and effectiveness. The result of this learning experience led to a greater comprehension about the use of fallacies, favorable perceptions about the use of games for educational experience, and a deeper reflection about social intelligence in the students.
History of scholarship and learning. The humanities, Social Sciences
In recent years, the essential roles of digital payment have gradually emerged. However, current research on digital payment adoption models rarely incorporates the outcomes of digital payment, and it also gives less consideration to rural residents. Considering these two issues as a research gap, this article establishes a unified digital payment use and credit availability model by extending the UTAUT2 framework in two aspects and applying it to China’s rural residents. The first extension is to add credit availability as an outcome variable of digital payment use. The second is to add two factors important to farmers, perceived riskiness and innovativeness, as constructs. Structural equation modeling is employed to analyze data collected from nearly 500 Chinese rural residents. The results show that almost 90% of rural residents have used digital payment. However, only a low proportion use it for many purposes or frequently. Rural residents’ digital payment use can increase credit availability. Performance expectancy, effort expectancy, social influence, price value, and innovativeness significantly and positively affect digital payment intention and behavior, thereby indirectly improving credit availability. However, the perceived risk does not influence digital payment adoption, possibly due to effective protective behaviors. In addition to the indirect effect, the results show that innovativeness also has a direct impact on credit availability. JEL Classification : M15, G21.
History of scholarship and learning. The humanities, Social Sciences
Bulbul Ahamed, Mohammad Rashed Hasan Polas, Ahmed Imran Kabir
et al.
The purpose of the study is to examine how cybersecurity knowledge, password security, and self-perception of skill affect cybersecurity awareness issues via the mediating lens of cybersecurity attitude among university students in Bangladesh. A sample of 430 university students from two public and three private universities provided the data in Dhaka, Bangladesh. An approach known as stratified random sampling was used in this cross-sectional study. The positivist approach was used, and a hypothetical statistical induction technique was used. The research constructs, which were adopted from earlier studies, were measured using scales that had undergone validation. Smart PLS-SEM 3.3.9 was used to quantitatively analyze the data. The results indicated a positive and significant association between cybersecurity knowledge and password security with cybersecurity awareness. No conventional association was found between self-perception of skills and cybersecurity awareness. Moreover, the data analysis confirmed that cybersecurity attitudemediates the relationship between cybersecurity knowledge, password security and self-perception of skills with cybersecurity awareness. This study implies that more effort needs to be put into informing the general people likely students about cybersecurity and ethical internet use. Furthermore, the main contribution of this study is to emphasize the need of raising cybersecurity awareness among students.
History of scholarship and learning. The humanities, Social Sciences
Tulay Ilhan-Nas, Aysegul Saglam, Tarhan Okan
et al.
Industry 4.0, whose effects have been more and more noticeable in recent years, and the digital change it brings call for a new educational model that aligns university instructional processes and curricula with the demands of business. This approach, known as University 4.0, intends to promote more technology-based applications, the power of innovation, the training of skilled specialists to satisfy industrial needs, and the development of competences that can achieve multidisciplinary projects. It is of utmost importance to adapt late-industrializing nations’ educational systems, such as Turkey’s, to this change process. The aim of this research, which takes the Turkish context as a reference, is to reveal how the curricula and educational processes of business schools should be shaped with a perspective that takes into account the human resources requirements of industry 4.0. A decision-making technique that integrates QFD, Delphi, and DEMATEL methods was employed in the study. The results clearly show that the most important expectation for social stakeholders is to provide university-industry cooperation. However, some of the most pressing problems include applied learning through collaborative educational experiences, teamwork skills, changing roles and competencies of academics. The findings point out that the educational activities of business schools must be linked with industry expectations through technology-based training activities and teamwork.
History of scholarship and learning. The humanities, Social Sciences
Small mining towns are often single-industry towns that turn to ghost towns or face negative socio-economic impacts upon mine closure. This study qualitatively explores the roles that mining companies and other key stakeholders (should) play in the development of local economies of the small mining communities to bring about economic sustainability, employing a constant comparative analysis. A small mining town in South Africa is the case study. Legislative and policy frameworks were scrutinized for their effectiveness in promoting economic sustainability. In-depth interviews with key stakeholders were carried out. Key factors limiting the effective implementation of developmental strategies were also explored. The study finds that weak community involvement, lack of trust, poor collaboration, poor municipal capacity, and legislation and policy flaws impact economic sustainability. Sustainable local economic development efforts are reported though. However, such efforts are insufficient because of the legislation and policy frameworks that are promoting short-term growth. Also, the town’s overdependence on the mining company, local government not optimally fulfilling their roles and responsibilities, and minimal community members’ participation on local economic development are other hindrances. However, the fact that the mining company and local municipality do acknowledge the shortcomings in their efforts towards promoting economic sustainability is a promise in minimizing the socio-economic effects of mine closures.
History of scholarship and learning. The humanities, Social Sciences
Based on the Internet Finance development index developed by Peking University in investigating 31 provinces, we use social network analysis to investigate the spatial correlation and influencing factors of the Internet Finance of China’s provinces. The research shows that the spatial correlation of Internet Finance of China’s provinces has significant characteristics of a regional gradient. The correlation among their correlation is moderate, more closely related, smooth, or weak. All provinces gather to form four plates; they play their respective functional advantages and have different statuses, functions, and roles. The differences between provinces in industrial structure, degree of marketization, infrastructure, degree of informatization, and geographical distance significantly affect the spatial correlation of Internet finance development in China, and the degree of influence decreases in turn. Our results might positively affect policymakers in promoting the coordinated development of regional Internet finance in China.
History of scholarship and learning. The humanities, Social Sciences
Shahriar Noroozizadeh, Jeremy C. Weiss, George H. Chen
We consider the problem of predicting how the likelihood of an outcome of interest for a patient changes over time as we observe more of the patient data. To solve this problem, we propose a supervised contrastive learning framework that learns an embedding representation for each time step of a patient time series. Our framework learns the embedding space to have the following properties: (1) nearby points in the embedding space have similar predicted class probabilities, (2) adjacent time steps of the same time series map to nearby points in the embedding space, and (3) time steps with very different raw feature vectors map to far apart regions of the embedding space. To achieve property (3), we employ a nearest neighbor pairing mechanism in the raw feature space. This mechanism also serves as an alternative to data augmentation, a key ingredient of contrastive learning, which lacks a standard procedure that is adequately realistic for clinical tabular data, to our knowledge. We demonstrate that our approach outperforms state-of-the-art baselines in predicting mortality of septic patients (MIMIC-III dataset) and tracking progression of cognitive impairment (ADNI dataset). Our method also consistently recovers the correct synthetic dataset embedding structure across experiments, a feat not achieved by baselines. Our ablation experiments show the pivotal role of our nearest neighbor pairing.
Data science has become increasingly essential for the production of official statistics, as it enables the automated collection, processing, and analysis of large amounts of data. With such data science practices in place, it enables more timely, more insightful and more flexible reporting. However, the quality and integrity of data-science-driven statistics rely on the accuracy and reliability of the data sources and the machine learning techniques that support them. In particular, changes in data sources are inevitable to occur and pose significant risks that are crucial to address in the context of machine learning for official statistics. This paper gives an overview of the main risks, liabilities, and uncertainties associated with changing data sources in the context of machine learning for official statistics. We provide a checklist of the most prevalent origins and causes of changing data sources; not only on a technical level but also regarding ownership, ethics, regulation, and public perception. Next, we highlight the repercussions of changing data sources on statistical reporting. These include technical effects such as concept drift, bias, availability, validity, accuracy and completeness, but also the neutrality and potential discontinuation of the statistical offering. We offer a few important precautionary measures, such as enhancing robustness in both data sourcing and statistical techniques, and thorough monitoring. In doing so, machine learning-based official statistics can maintain integrity, reliability, consistency, and relevance in policy-making, decision-making, and public discourse.
A fines del siglo XIX, el sistema educativo en Argentina adoptó la idea de una lengua única. La conformación del Estado nación se apoyó en el modelo europeo, buscando construir una nacionalidad para lograr la homogeneización de los sujetos y la lengua fue utilizada como un instrumento necesario para tal fin. Para ello, diversas lenguas e identidades fueron negadas, excluidas y expulsadas hacia el otro lado de la frontera que conforma una mismidad (la «nación») construida a partir de modelos occidentales. En este sentido, las políticas que instauraron y difundieron al castellano como idioma oficial, y el establecimiento de la obligatoriedad de su enseñanza, generaron y perpetuaron desigualdad y discriminación social. Muchos intelectuales de la época han polemizado y debatido en torno al lenguaje que se debía hablar en el nuevo Estado.
En este artículo se realiza la descripción y el análisis de la concepción sobre la lengua, específicamente sobre el castellano, y sobre su enseñanza por parte del intelectual Ricardo Rojas. Por otro lado, se analizan las representaciones que este expone sobre los pueblos indígenas y sobre sus lenguas. Para ello, utilizamos como fuentes La restauración nacionalista: Informe sobre educación, escrita en 1909, y Blasón de plata, publicada en 1910.
Rojas postula buscar el fundamento de la conciencia nacional en España y en los pueblos indígenas, reconociéndolos como un eslabón más de nuestro pasado argentino. De esta forma, propone el estudio de sus lenguas con el propósito de enriquecer las investigaciones sobre la tradición. Esto implicó polémicos debates contra los discursos intelectuales hegemónicos de principios del siglo XX que se encuadraban en el pensamiento sarmientino y que alzaban las banderas del positivismo biologicista. No obstante, su consideración sobre los indígenas es la de sujetos primitivos, del pasado y ligados a la naturaleza más que a la civilización.
Social Sciences, History of scholarship and learning. The humanities
This study investigated the effect of media and information literacy (MIL) on the ability to identify fake news, disinformation and misinformation, and sharing intentions. The experimental approach was selected to study both the control group and experimental group made up of a total of 187 respondents. Comparative analysis of the two groups revealed that although more respondents in the experimental group were able to identify the inauthenticity of information presented to them, some of the respondents in the control group were also able to do the same, even though they did not receive MIL training. Conversely, some respondents in the experimental group, even though they were trained in MIL, could not determine the inauthenticity of information, possibly because the one-off training given to them did not allow them to assimilate all the information in one sitting. Nonetheless, the results of the bivariate correlation computation showed that MIL trained respondents were more likely to determine authenticity or otherwise of information and less likely to share inaccurate stories. This means that when MIL increases, sharing of fake news decreases. This is yet another evidence that MIL enables information consumers to make informed judgments about quality information. It is recommended that MIL is incorporated into mainstream educational modules and consistently revised to reflect the demands of the times. MIL programs must also consider how to effectively reach those without formal education. Actors within the information, communications, and media ecology must contribute to their quota in making information consumers more discerning with the right MIL sensitisation.
A key challenge for machine intelligence is to learn new visual concepts without forgetting the previously acquired knowledge. Continual learning is aimed towards addressing this challenge. However, there is a gap between existing supervised continual learning and human-like intelligence, where human is able to learn from both labeled and unlabeled data. How unlabeled data affects learning and catastrophic forgetting in the continual learning process remains unknown. To explore these issues, we formulate a new semi-supervised continual learning method, which can be generically applied to existing continual learning models. Specifically, a novel gradient learner learns from labeled data to predict gradients on unlabeled data. Hence, the unlabeled data could fit into the supervised continual learning method. Different from conventional semi-supervised settings, we do not hypothesize that the underlying classes, which are associated to the unlabeled data, are known to the learning process. In other words, the unlabeled data could be very distinct from the labeled data. We evaluate the proposed method on mainstream continual learning, adversarial continual learning, and semi-supervised learning tasks. The proposed method achieves state-of-the-art performance on classification accuracy and backward transfer in the continual learning setting while achieving desired performance on classification accuracy in the semi-supervised learning setting. This implies that the unlabeled images can enhance the generalizability of continual learning models on the predictive ability on unseen data and significantly alleviate catastrophic forgetting. The code is available at \url{https://github.com/luoyan407/grad_prediction.git}.
La población infantil en algunas zonas rurales, se presenta como un sector con una potencial necesidad de realizarle estudios que contemplen la calidad de vida desde dimensiones subjetivas, debido a que son zonas consideradas como vulnerables por la alta precariedad y marginación, factores que pueden comprometer un óptimo desarrollo biopsicosocial de los niños. Abordar estudios desde una mirada cualitativa, permitirá no solo la visibilización de los niños como grupo social, sino que además aportará información desde su propia perspectiva de experimentar, interpretar y configurar el mundo que los rodea, de esta manera la información encontrada podrá ser de utilidad para fundamentar políticas públicas, programas y acciones para promover el bienestar y la calidad de vida. El siguiente escrito expone la problemática social y económica en la que se desenvuelven los niños hijos de jornaleros agrícolas, se hace una asociación de cómo estas condiciones desfavorables de carencia y marginación pueden incidir en el bienestar y la calidad de vida e influir a su vez, en la aparición de alteraciones psicológicas como la sintomatología depresiva. Se enmarca la importancia de profundizar en el estudio de la calidad de vida desde la metodología cualitativa, la cual sitúa al niño como un ser activo de la sociedad, que tiene la libertad de construir la realidad desde su mirada y con base en sus experiencias.
History of scholarship and learning. The humanities, Social sciences (General)
MONEDELE OLANDEZE ÎN PRINCIPATUL MOLDOVEI (1711-1859)1 Absența unei monede naționale în Principatul Moldovei a permis pătrunderea diverselor tipuri de monede străine în acest spațiu. Importanța monedelor olandeze în Principatul Moldovei pe parcursul secolelor XVIII și XIX este una esențială, fapt confirmat atât de sursele scrise, cât și de descoperirile monetare. Atestarea monedelor olandeze sub formă de descoperiri monetare, pe o zonă extinsă a Principatului Moldovei, dar și menționarea lor în diferite surse scrise, confirmă răspândirea și importanța monedei olandeze în circulația monetară din acest teritoriu. Prezența talerului și a ducatului în circulația monetară a fost atât de semnificativă, încât se va produce situația în care aceste monede, deși internaționale, vor deține rolul de monede oficiale. Cuvinte-cheie: Principatul Moldovei, monedă, circulaţie monetară, tezaur, taler, ducat.
History of scholarship and learning. The humanities
Article is devoted to insufficiently studied aspects of the early history of Cossacks villages of Cuban line regiment during the first ten years of its service on the new Caucasian border. On the basis of archival and published documents, the author studies the religious features of migrants, the statistics of their migrations and describes the first officers’ staff of the Cuban regiment. The author offers solutions to the research problems that have not been resolved by the previous imperial and Soviet regional historiography.
Law, History of scholarship and learning. The humanities
Krishna Pillutla, Yassine Laguel, Jérôme Malick
et al.
We present a federated learning framework that is designed to robustly deliver good predictive performance across individual clients with heterogeneous data. The proposed approach hinges upon a superquantile-based learning objective that captures the tail statistics of the error distribution over heterogeneous clients. We present a stochastic training algorithm that interleaves differentially private client filtering with federated averaging steps. We prove finite time convergence guarantees for the algorithm: $O(1/\sqrt{T})$ in the nonconvex case in $T$ communication rounds and $O(\exp(-T/κ^{3/2}) + κ/T)$ in the strongly convex case with local condition number $κ$. Experimental results on benchmark datasets for federated learning demonstrate that our approach is competitive with classical ones in terms of average error and outperforms them in terms of tail statistics of the error.