Search Results (4,394)

Plain Language seeks to enhance the clarity and comprehensibility of legal and administrative communication; while Natural Language Processing (NLP) offers promising tools for assessing text complexity, most Plain Language classification studies focus exclusively on English, leaving low-resource languages underexplored. This study presents the first systematic evaluation of transformer-based models for sentence-level Plain Language classification in Hungarian tax administrative texts. We benchmarked zero-shot prompting with GPT-4o against fine-tuned open-weight and proprietary models, including huBERT, XLM-RoBERTa, GPT-4o-mini, and Gemini 1.0 Pro, and contextualized these results against previously established lightweight machine learning baselines based on term frequency-inverse document frequency with a support vector machine (TF-IDF + SVM) and fastText. To address data scarcity, we applied translation-based data augmentation using parallel Hungarian–English corpora. The best-performing model achieved a macro-average F1-score of 0.79. Mid-sized models also delivered competitive results, combining accuracy with feasible inference speed and deployment flexibility. Beyond classification performance, we conducted local and aggregated interpretability analysis based on Shapley-values to identify linguistic patterns influencing model decisions. This revealed alignment with known Plain Language features, such as nominalizations and syntactic complexity, as well as biases introduced by frequent domain-specific terms. Our findings demonstrate that Plain Language classifiers can be effectively adapted to low-resource legal–administrative domains. The results support the development of real-time feedback tools that promote linguistic accessibility and contribute to the broader goal of Access to Justice. Full article

Within the framework of argument realization theory, Semantic Role Lists, Participant Roles, and Predicate–Argument Structures represent competing models of verb semantics, each yielding distinct predictions regarding mental representation and processing complexity. This study formalizes these perspectives into two competing accounts: the Role-Quantity Hypothesis (H1), which posits that processing load is driven by the number of event participants, and the Structure-Quantity Hypothesis (H2), which attributes complexity to the multiplicity of syntactic templates. To evaluate these hypotheses, a lexical decision task was conducted on Chinese verbs. The results revealed a significant main effect of role quantity: two-role verbs elicited longer reaction times and lower accuracy than one-role verbs. Conversely, no significant differences were found between one-structure and two-structure verbs. These findings provide robust empirical support for H1, indicating that role-based representations possess greater psychological reality in the Chinese mental lexicon. We argue that for an isolating language like Chinese, verb processing is primarily event-driven, where role information serves as a predictive heuristic during early lexical access. This study offers new insights into the role-based nature of Chinese verb representation, its psychological reality in real-time processing, and the value of integrating argument realization theory with experimental psycholinguistics. Full article

Large language models (LLMs) are increasingly deployed in multilingual settings, yet their safety behavior under Turkish harmful prompts and prompt injection attempts remains insufficiently characterized. This study evaluates the adversarial robustness of 55 open- and closed-source LLMs under paired Turkish and English harmful prompt conditions. We constructed a benchmark of 790 Turkish adversarial prompts, translated the prompts into English for cross-lingual comparison, and applied both prompt sets to the model pool. Model responses were labeled as harmful, harmless, or hallucinatory, and safety was analyzed using safety scores, Turkish–English ranking differences, and inter-rater reliability based on Fleiss’ kappa. The results reveal substantial variation across models. Closed-source systems generally achieved higher safety scores and stronger filtering behavior, whereas open-source and Turkish-oriented models showed a wider performance distribution. GPT-5.4 ranked first in the Turkish tests with a 99.37% safety score but decreased to 96.71% in the English tests, while Qwen3.5:27B ranked first in English with 97.47%. These differences suggest that safety mechanisms are not fully language-invariant. Hallucination also emerged as a distinct safety risk, particularly in Turkish evaluations. The findings indicate that Turkish LLM safety cannot be inferred from general model capability alone and should be assessed through language-specific, culturally aware, and continuously updated adversarial benchmarks. Full article

Social media has become a critical channel for public health communication during the COVID-19 pandemic, yet how official health messaging aligns with broader public discourse remains insufficiently understood. This study develops an end-to-end infoveillance framework to examine the dynamic relationship between Centers for Disease Control and Prevention (CDC) communications and general public discourse on social media. We analyzed 17,524 CDC tweets and 67,895 public discourse tweets. Biterm Topic Model (BTM) was used to extract topics from each corpus, and a novel topic consistency scoring system integrating cosine similarity with daily public topic prominence was developed to quantify temporal alignment between official health communication and public discourse. Two complementary sentiment measures were incorporated: expected sentiment (average emotional tone) and net sentiment (overall emotional intensity). Temporal relationships were examined using autoregressive integrated moving average with exogenous variables (ARIMAX) models. Results show that topic alignment increased over time across CDC topics, while expected sentiment remained consistently negative. Higher alignment was associated with immediate and delayed changes in expected sentiment and stronger emotional intensity in net sentiment based on ARIMAX results. These findings suggest that topic alignment reflects public attention rather than agreement with official communications and is associated with more negative emotional responses. This framework provides a scalable, generalizable approach to investigating and evaluating public engagement with official health communication. Full article

In the context of industrial digitalization and the widespread adoption of process automation systems, Knowledge Management Systems (KMS) play a key role in providing operational personnel with up-to-date instructions and regulations. However, the inherent ambiguity of natural language in technical documentation remains a serious obstacle, leading to incorrect operator actions, process deviations, and increased safety risks. This article investigates the integration of Large Language Models (LLMs) into KMS and its impact on user experience and human–machine interaction in industrial automation environments. A method called Semantic Latent Choice Detection is presented, designed to systematically identify interpretation ambiguities in process instructions and operator commands. Unlike existing approaches that require access to the internal model architecture (“white box”) or token-level logits, the proposed method is logit-free and operates with closed commercial LLMs (“black box”) via standard API interfaces. The method analyzes the semantic similarity of binary text blocks and polysemous terms within the context of a specific technological process. Using a metallurgical production case study, we demonstrate how the system detects hidden semantic collisions (e.g., the difference between “adding ferroalloys into the ladle” and “feeding ferroalloys onto the conveyor”) that are missed by traditional rule-based validation methods. Instead of arbitrarily selecting an interpretation, the system initiates a clarification request to the human operator, thereby reducing cognitive load, preventing erroneous automated decisions, and increasing trust in the KMS. An empirical evaluation conducted in a real-world industrial setting (unit control rooms and dispatch centers) shows a statistically significant reduction in errors related to misinterpretation of process regulations. The article contributes to the fields of automation engineering, knowledge management, and human-centered automation by proposing a novel method for validating operational instructions in high-risk industrial environments. Full article

Treatment of non-small cell lung cancer (NSCLC) has been transformed by immunotherapy and targeted therapies. We aimed to characterize clinical features, treatment patterns, and healthcare resource use in patients with early and locally advanced NSCLC before incorporation of these therapies. This retrospective observational study included adults diagnosed with stage I–III NSCLC at four Spanish hospitals between 2014 and 2018, with follow-up until 2021, using artificial intelligence to extract data from electronic health records. A total of 951 patients were included (34.7% stage I, 16.7% stage II, 48.6% stage III), with a median age of 66 years and 31.9% female. Surgery was performed in 78.5% of stage I, 74.8% of stage II, and 35.5% of stage III patients. Among surgical patients, 62.5% received adjuvant chemo- and/or radiotherapy, 20.8% neoadjuvant therapy, and 15.7% both; among non-surgical patients, chemoradiotherapy was the most common treatment (50.4%). Beyond hospitalization, outpatient visits were the most frequently used healthcare resource. These findings provide a historical benchmark of NSCLC care before introduction of immunotherapy and targeted therapies in these settings, highlighting treatment variability and the need for earlier diagnosis, structured treatment pathways, and multidisciplinary management. Full article

Background: Psychological distress, particularly anxiety and depression, is highly prevalent among cancer patients, and is associated with impaired quality of life, reduced treatment adherence, and increased mortality risk. Standardized screening instruments, such as the Hospital Anxiety and Depression Scale (HADS), are effective, but face implementation barriers in busy oncology outpatient settings. This cross-sectional study investigated whether BERT-based Natural Language Processing (NLP) analysis of brief patient-generated free texts would correlate with HADS scores in a consecutive cohort of cancer outpatients. Material and Methods: A total of 165 consecutive adult cancer outpatients were enrolled at a tertiary oncology center in Turkey. All participants completed the HADS questionnaire and were asked to write freely about their current emotional state in Turkish. Patient-generated texts were analyzed using a pre-trained Turkish BERT model to derive a continuous BERT Sentiment Score (BSS) and a categorical BERT Sentiment Cluster (BSC) via unsupervised hierarchical clustering. Univariate and multivariate linear regression analyses were performed to examine associations between clinical, demographic, and NLP-derived variables and the logarithmically transformed HADS score. Results: The mean total HADS score was 10.46 (range, 0–33), consistent with a moderate level of psychological distress. In multivariate analysis, two variables were independently associated with HADS scores: female sex (β = 0.20, t = 2.14, p = 0.034), associated with higher HADS scores, and BERT Sentiment Score (BSS) (β = −0.18, t = −2.43, p = 0.016), with higher values corresponding to lower HADS scores. Hierarchical clustering identified two distinct thematic groups: ‘Coping and Fighting Spirit’ (74%), and ‘Hope and Negative Feelings’ (26%); however, cluster membership (BSC) was not independently associated with HADS scores (β = −0.02, p = 0.789). Clinical variables, including cancer stage, diagnosis type, treatment status, and time since diagnosis, also were not independently associated with HADS scores. Conclusions: BERT-based sentiment analysis of brief patient-generated free texts yielded a continuous measure that independently correlated with HADS scores in cancer outpatients, alongside female sex. These findings provide proof-of-concept evidence that NLP-derived sentiment scoring may offer a practical, scalable, and complementary approach to standardized psychological screening in routine oncology care. Full article

Construction contracts are vital for governing responsibilities in large-scale infrastructure projects, but their increasing complexity often leads to interpretation difficulties, disputes, and delays. Despite advances in natural language processing (NLP), automated analysis of construction contract clauses remains limited in project management. This study proposes a text classification framework integrating transformer-based contextual embeddings (BERT, ALBERT, RoBERTa, and DistilBERT) with machine learning and deep learning models (RNN, GRU, and LSTM) to analyze FIDIC and JCT contract provisions. Two multi-class classification tasks were defined: Dataset 1 (DS1) focusing on obligations, operational actions, optional provisions, general statements, and Dataset 2 (DS2) covering cost, quality, and time dimensions. Experimental results show that deep learning models consistently outperform traditional machine learning algorithms. Specifically, LSTM combined with RoBERTa and DistilBERT achieved the highest accuracy levels of 98.06% and 98.33% for DS1. The framework may support transparent contract governance by enabling faster and more consistent identification of contractual clauses. From a sustainability perspective, the findings suggest potential process-level contributions to economic efficiency, administrative workload reduction, and decision-making support throughout the project lifecycle. Full article

The rapid expansion of the Internet has enabled large-scale information sharing while also increasing the risk of sensitive information leakage. Authentication secrets, including passwords and API keys, may be unintentionally exposed in publicly accessible environments such as web pages, network packets, and code-sharing platforms when they are mishandled by developers or operators. Such leakage allows attackers to abuse third-party authentication services and may lead to unauthorized access, fraud, or broader compromise. Therefore, timely and accurate detection of sensitive information in network data is essential for reducing security risks. This paper presents SenScanner, an artificial intelligence-based model for automatically identifying password-related secrets in mixed text. By combining natural language processing and machine learning techniques, SenScanner detects leaked password-related sensitive information across heterogeneous textual contexts. Experimental results on 2000 public data samples show that SenScanner achieves superior precision, recall, and F1-score, demonstrating its effectiveness in reducing false positives and manual review effort. Full article

This study investigates Bitcoin price forecasting using integrated sentiment analysis and gradient boosting within digital financial ecosystems. Two XGBoost models were developed using sentiment scores derived from Bitcoin news (2021–2024) and technical indicators, including GARCH-estimated volatility, Bollinger Bands, MACD, and RSI. The analysis uses 1042 daily Bitcoin observations and 10,025 sentiment records. Two model configurations were evaluated: one using only technical indicators and another incorporating daily aggregated sentiment scores. Model performance was assessed using Diebold–Mariano tests with Newey–West HAC variance estimation and walk-forward validation across 40 rolling windows. Contrary to expectations, sentiment features provided no statistically significant improvement over the technical-only model (p = 0.4888). Both models achieved identical test performance (R² = −0.16%). Walk-forward validation revealed substantial temporal instability (Mean R² = −126.30%, Std = 233.05%), highlighting the challenges of forecasting daily Bitcoin returns. Nevertheless, both XGBoost models significantly outperformed the random walk benchmark (DM statistic = −8.58, p < 0.0001), indicating that technical indicators capture exploitable market structure despite limited predictive accuracy for practical trading. These findings support the efficient market hypothesis and have implications for digital financial ecosystems integrating multimodal information. Full article

This study investigates a target-based sentiment analysis approach on Turkish texts and examines how lexicon-based methods vary depending on language compatibility and translation strategies. The main objective is to accurately identify target-oriented expressions and to compare the performance of different sentiment lexicons within this context. For this purpose, Turkish user reviews obtained from the Turkish school review and evaluation platform were analysed using three lexicon configurations: SentiWordNet applied in its original English form with target-related term translation (SentiWordNet-EN), its fully Turkish-translated version (SentiWordNet-TR), and a native Turkish resource (SentiTurkNet). SentiTurkNet achieved the highest weighted average F1-score of 0.887 (positive-class F1: 0.926; negative-class F1: 0.760), followed by SentiWordNet-EN with a weighted average F1-score of 0.856 (positive-class F1: 0.898; negative-class F1: 0.720), and SentiWordNet-TR with a weighted average F1-score of 0.824 (positive-class F1: 0.868; negative-class F1: 0.679). One of the most significant findings is that using SentiWordNet in its original English form yields better results than the fully translated version, suggesting that the translation process leads to sentiment loss due to the incomplete preservation of sentiment intensity and contextual meaning. These findings carry important implications for sentiment analysis in low-resource languages: where comprehensive native lexicons are unavailable, translating only target-related terms into a language with richer sentiment resources can be more effective than directly translating the entire lexicon. Therefore, it is concluded that in target-based sentiment analysis, not only language compatibility but also the chosen translation strategy plays a critical role. Full article

Virtual influencers are increasingly used in influencer marketing, yet it remains unclear whether trust generated by an artificial persona can be transferred to endorsed brands in the same way as trust generated by human influencers. This study examines artificial intelligence (AI)-mediated trust transfer among female consumers by comparing virtual and human influencers across four social media platforms. Drawing on source credibility, parasocial interaction, social presence, trust transfer, and brand attachment perspectives, we propose that influencer type is associated with brand outcomes through three observable social-cue pathways: perceived authenticity, parasocial interaction, and social presence. These cues are expected to be associated with influencer trust, which is then associated with brand trust, brand attachment, purchase intention, and recommendation intention. Using 78,432 female consumer comments from Xiaohongshu, Instagram, Weibo, and Douyin, matched across 12 virtual–human-influencer pairs, we construct text-derived linguistic indicators—proxies rather than validated psychometric constructs—through keyword dictionaries, sentiment classification, and standardized composite scoring. The results show that human influencers are associated with higher perceived authenticity, parasocial interaction, and social presence than virtual influencers. The strongest association in the model is the trust-transfer path from influencer trust to brand trust, and the indirect path is associated with a substantial share of the observed covariation between influencer type and brand attachment. Product type further qualifies these patterns: the human-influencer advantage is stronger for hedonic products than for utilitarian products. These findings suggest that virtual influencers should not be understood as universal substitutes for human influencers. Instead, their effectiveness depends on whether AI-mediated personas can generate the social and authenticity cues that the literature associates with trust transfer in a given product context. Full article

Paprika is a spice obtained from the dehydration and grinding of red pepper fruits, primarily from the Capsicum annuum species. Its etymology comes from Slavic Balkanian languages and was adopted in Hungarian. The crop originated in America, where it was domesticated by pre-Columbian civilizations over 6000 years ago (specifically in present-day Mexico) for medicinal and culinary purposes. Following the Spanish arrival in the Americas in the 15th century, pepper was introduced first in Spain (Sevilla, Extremadura and Murcia) and later in the rest of the Old World. The agroclimatic conditions of different Mediterranean regions made it an essential crop, turning these regions into centers of production and giving this spice a sense of cultural identity. The purpose of this study lies in the technological and nutritional significance of paprika in the modern food industry, where it is demanded as a natural colorant, preservative and source of bioactive compounds, such as antioxidants and carotenoids. Despite its prevalence, the existing literature is often fragmented into specific disciplines. This article distinguishes itself by proposing a holistic approach expanding the study from its historical evolution to its socioeconomic impact, including its agronomic characteristics and industrial-scale production. It is recommended that the research community and producers focus on the sustainability of processing methods while preserving cultural authenticity, ensuring the preservation of the functional and culinary relevance of this spice. Full article

The evolution of smart educational platforms has been significantly driven by the global development of ML (Machine Learning), particularly through the application of NLP (Natural Language Processing) to personalize student learning experiences. However, high schools in Gauteng Province face significant challenges in adopting smart educational portals. This is due to inadequate accessibility to resources, insufficient personalization mechanisms, and low student engagement frameworks caused by the failure to adapt to proven teaching methods. Despite the successful adoption of NLP and ML in global case studies and implementations, local gaps persist. Specifically, pretrained LLMs (Large Language Models) such as BERT (Bidirectional Encoder Representations from Transformers) require fine-tuning for African languages, yet little testing of these tools in Gauteng’s public high schools has been done. This review uses a structured literature review methodology, which examines peer-reviewed studies, case reports, and technical documents published between 2014–2026. Findings indicate that multilingual NLP resources for South African languages remain severely underdeveloped. Furthermore, it demonstrates that current smart-learning portals lack inclusive design adjustments for multilingual and low-resource contexts. Based on these findings, the paper recommends strategies for enhancing accessibility, personalization, and engagement. This includes the development of multilingual NLP resources, optimization of ML architectures for constrained infrastructure, and context-aware pedagogical adaptations. The review follows a two-layer design, consisting of (i) a global systematic synthesis of NLP and ML applications in education, and (ii) a contextual interpretation of these findings for Gauteng high schools using regional policy documents, infrastructure reports, and educational statistics. The conclusions pertain to implications and recommendations for Gauteng high schools, rather than evaluation of an existing local portal. This paper highlights the potential of NLP and ML to transform education in Gauteng but highlights the urgency of localized research and ML implementation. Full article

Arabic multi-label emotion detection (MLED) in social media remains challenging because dialectal variation, implicit affective cues, and polarity-opposed emotions may occur within the same post. Existing Arabic MLED studies have mainly emphasized thresholded predictive performance, with limited attention to whether model confidence remains reliable under emotionally conflicting conditions. In this study, we propose CONCORD-Emo (CONflict-aware Compositional Representation for Emotion Detection), a reliability-aware framework for Arabic MLED. The framework adopts established label-wise attention, mixture-of-experts routing, Monte Carlo (MC) dropout, and post hoc temperature scaling as supporting mechanisms, while its architecture-level contribution is the conflict-conditioned integration of a residual global anchor with a conflict-aware fusion gate supervised by an automatically derived polarity-conflict target. We evaluated the framework on three Arabic benchmarks: SemEval-2018-Ar, ExaAEC, and SemEval-2025-Arq using predictive and reliability-oriented criteria. CONCORD-Emo remains competitive with strong MARBERT-based baselines. On SemEval-2025-Arq, it attains point estimates of 0.471 for Jaccard, 0.606 for micro-F1, and 0.582 for macro-F1. Paired bootstrap confidence intervals show that most predictive differences include zero, whereas the lower Expected Calibration Error and Brier scores on SemEval-2018-Ar and ExaAEC are consistently supported relative to the controlled baselines. Conflict-conditioned analysis shows that polarity-conflict instances yield lower predictive performance and higher Brier scores than blended-emotion instances. Taken together, these results support a reliability-aware evaluation of Arabic MLED in which polarity conflict, calibration, uncertainty estimation, and selective prediction are examined alongside predictive performance. Full article