Accurate forecasting of bus ridership (passengers numbers) is crucial for efficient management and optimization of public transport systems. Traditional forecasting models often fail to capture the unique and localized dynamics of different urban areas by treating the entire city as a single, homogeneous region. This paper introduces a novel framework that enhances bus ridership prediction by integrating a spatial clustering methodology with multi-dimensional feature analysis. The proposed framework utilizes a diverse set of data, including bus ridership data (by route number, time, and bus stop) complemented by a variety of open source data, such as spatial features (e.g., attractive destinations), meteorological conditions (e.g., temperature, rainfall), and temporal patterns (e.g., time of day, day of week). By clustering the urban area into distinct regions, based on the principle that bus stops in close proximity share similar ridership characteristics, a separate local forecasting model is trained for each of these clusters. This localized approach demonstrates an accuracy comparable to that of global models. The findings suggest that a spatially-aware, localized modeling strategy is effective for public transport prediction, paving the way for more targeted and efficient service improvements.

Radio broadcasting remains a dominant medium of communication, reaching 82 The dataset comprises filtered recordings from 74 US news and talk radio stations, collected over a 24-hour period and yielding more than 1,400 hours of content. We analyzed the data using VANPY, an in-house voice-analysis framework that combines multi-channel radio recording with AI-based speaker diarization, gender classification, speech-to-text transcription, and topic analysis. The results revealed consistent gender differences in allocated broadcast time, with male speakers accounting for 77 Beyond these findings on gender dynamics in broadcast media, VANPY is publicly available and provides a systematic approach for analyzing large-scale audio data. It can be applied not only to radio broadcasts, but also to other audio domains such as human-computer interaction, corporate communication, and security applications.

Scientific papers are a primary source of knowledge for research and development, yet the information they contain is highly unstructured as it being spread across text, tables, and figures in inconsistent formats. Structuring this dispersed information into a coherent database is a central informatics challenge. In this work, we present a generic artificial intelligence (AI)-driven methodology to automate the construction of domain-specific databases from scientific literature. The methodology includes three Large Language Model (LLM)-based stages: identifying domain-specific literature, classifying papers by relevance, and extracting the scientific data they contain. We demonstrate and validate this approach in a case study related to chemical reactions associated with environmental issues by constructing a catalytic performance database for the CO hydrogenation process. Data from over 1,000 disparate papers were extracted and transformed into a tabular database, containing over 9,600 entries. The extraction performance proved highly effective, achieving an F1 score of >99% for text and tables, and 85.8% for figures. This methodology can be easily adapted and implemented in other fields.

Large Language Models (LLMs) are increasingly shaping how we consume information and form our worldview. This raises concerns beyond bias in AI: do LLMs grasp the emotional nuances conveyed via textual framing? In this work, we empirically evaluate how well an array of LLMs aligns with human emotional perception. Considering news headlines covering political and geopolitical conflicts, both human participants (n = 3011, a representative sample of the U.K. adult population, via a YouGov survey) and seven LLMs answered whether headlines evoked sympathy for a specified side in a conflict. We find that the correlation between AI and human evaluations varies across models, ranging from very high (0.789, GPT-5.2) to medium (0.4 ,Mistral Large 2512). Crucially, the leading models are broadly aligned with human judgments across all demographic subgroups, including age, gender, level of education, prior geopolitical knowledge, and participants' predispositions regarding the conflict, although there are statistically significant differences between groups. This research, with its robust design and large, demographically diverse dataset, offers the most comprehensive evaluation of LLMs' comprehension of news framing to date. Findings highlight an important, often-ignored aspect of differential alignment: even when aggregate performance is high, AI alignment is not universal – it may correspond differently with demographic features and cultural norms. Considering or ignoring the need for differential alignment may therefore have significant implications for the development of ethical and useful AI systems.

Software supply-chain attacks increasingly exploit an identity gap where compromised maintainer accounts authorize malicious changes. This work evaluates patch-level authorship verification as a behavioral defense layer, showing that stylometric analysis can operate not only on full source files but also on patch-level commits. We fine-tune a cross-modal transformer on more than 20 years of Linux kernel commit history to embed code diffs and commit messages into a unified stylometric space, achieving ROC AUC of 0.93 for open-world authorship verification. We then use these representations in a streaming anomaly detector suited to continuous integration and deployment (CI/CD) settings. We validate the pipeline on two retrospective supply-chain incidents involving different patch characteristics: the 2021 PHP backdoor and the 2026 ForceMemo/GlassWorm campaign. Without retraining, the proposed detector surfaces both PHP forged commits within approximately 1

Understanding how co-authors distribute credit is critical for accurately assessing scholarly collaboration. In this study, we uncover the implicit structures within scientific teamwork by systematically analyzing author contributions across a large corpus of research publications. We introduce a computational framework designed to convert free-text contribution statements into 14 standardized CRediT categories, identifying clear and consistent positional patterns in task assignments. By analyzing over 400,000 scientific articles from prominent sources such as PLOS One and Nature, we extracted and standardized more than 5.6 million author-task assignments corresponding to 1.58 million author mentions. Our analysis reveals substantial disparities in workload distribution. Notably, in small teams with three co-authors, the most engaged contributor performs over three times more tasks than the least engaged, a disparity that grows linearly with team size. This demonstrates a consistent pattern of central and peripheral roles within modern collaborative teams. Moreover, our analysis shows distinct positional biases in task allocation: technical responsibilities, such as software development and formal analysis, broadly fall to authors positioned earlier in the author list, whereas managerial tasks, including supervision and funding acquisition, increasingly concentrate among authors positioned toward the end. This gradient underscores a significant division of labor, where early-listed authors mainly undertake most hands-on activities. In contrast, senior authors mostly assume roles involving leadership and oversight. Our findings highlight the structured and hierarchical organization within scholarly collaborations, providing deeper insights into the specific roles and dynamics that govern academic teamwork

Embedding news articles is a crucial tool for various fields, including media bias detection, fake news identification, and news recommendation systems. However, existing news embedding methods are not optimized to capture the latent context of news events. Most embedding methods rely on full-text information, neglecting the generation of time-relevant embeddings. In this paper, we propose a novel, lightweight method that optimizes news embedding generation by focusing on entities and themes mentioned in articles and their historical connections to specific events. We suggest a three-stage method. First, we process and extract events, entities, and themes from news articles. Second, we generate periodic time embeddings for themes and entities by training time-separated GloVe models on current and historical data. Lastly, we concatenate the news embeddings generated by two distinct approaches: Smooth Inverse Frequency (SIF) for article-level vectors and Siamese Neural Networks for embeddings with nuanced event-related information. We leveraged over 850,000 news articles and one million events from the GDELT project to test and evaluate our method. We conducted a comparative analysis of different news embedding generation methods for validation. Our experiments demonstrate that our approach improves and outperforms state-of-the-art methods on shared event detection tasks.

Brain-Computer Interface (BCI) applications provide a direct way to map human brain activity onto the control of external devices, without a need for physical movements. These systems, crucial for medical applications and also useful for non-medical applications, predominantly use EEG signals recorded non-invasively, for system control, and require algorithms to translate signals into commands. Traditional BCI applications heavily depend on algorithms tailored to specific behavioral paradigms and on data collection using EEG systems with multiple channels. This complicates usability, comfort, and affordability. Moreover, the limited availability of extensive training datasets limits the development of robust models for classifiying collected data into behavioral intents. To address these challenges, we introduce an end-to-end EEG classification framework that employs a pre-trained Convolutional Neural Network (CNN) and a Transformer, initially designed for image processing, applied here for spatiotemporal representation of EEG data, and combined with a custom developed automated EEG channel selection algorithm to identify the most informative electrodes for the process, thus reducing data dimensionality, and easing subject comfort, along with improved classification performance of EEG data onto subjects intent. We evaluated our model using two benchmark datasets, the EEGmmidb and the OpenMIIR. We achieved superior performance compared to existing state-of-theart EEG classification methods, including the commonly used EEGnet. Our results indicate a classification accuracy improvement of 7% on OpenMIIR and 1% on EEGmmidb, reaching averages of 81% and 75%, respectively. Importantly, these improvements were obtained with fewer recording channels and less training data, demonstrating a framework that can support a more efficient approach to BCI tasks in terms of the amount of training data and the simplicity of the required hardware system needed for brain signals. This study not only advances the field of BCI but also suggests a scalable and more affordable framework for BCI applications. ### Competing Interest Statement The authors have declared no competing interest. ### Funding Statement This study did not receive any funding ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes All data produced are available online at openmiir and physionet

Keeping track of how lawmakers vote is essential for government transparency. While many parliamentary voting records are available online, they are often difficult to interpret, making it challenging to understand legislative behavior across parliaments and predict voting outcomes. Accurate prediction of votes has several potential benefits, from simplifying parliamentary work by filtering out bills with a low chance of passing to refining proposed legislation to increase its likelihood of approval. In this study, we leverage advanced machine learning and data analysis techniques to develop a comprehensive framework for predicting parliamentary voting outcomes across multiple legislatures. We introduce the Voting Prediction Framework (VPF) - a data-driven framework designed to forecast parliamentary voting outcomes at the individual legislator level and for entire bills. VPF consists of three key components: (1) Data Collection - gathering parliamentary voting records from multiple countries using APIs, web crawlers, and structured databases; (2) Parsing and Feature Integration - processing and enriching the data with meaningful features, such as legislator seniority, and content-based characteristics of a given bill; and (3) Prediction Models - using machine learning to forecast how each parliament member will vote and whether a bill is likely to pass. The framework will be open source, enabling anyone to use or modify the framework. To evaluate VPF, we analyzed over 5 million voting records from five countries - Canada, Israel, Tunisia, the United Kingdom and the USA. Our results show that VPF achieves up to 85% precision in predicting individual votes and up to 84% accuracy in predicting overall bill outcomes. These findings highlight VPF's potential as a valuable tool for political analysis, policy research, and enhancing public access to legislative decision-making.

Online social networks usage has increased significantly in the last decade and continues to grow in popularity. Multiple social platforms use volunteers as a central component. The behavior of volunteers in volunteer-based networks has been studied extensively in recent years. Here, we explore the development of volunteer-based social networks, primarily focusing on their key users' behaviors and activities. We developed two novel algorithms: the first reveals key user behavior patterns over time; the second utilizes machine learning methods to generate a forecasting model that can predict the future behavior of key users, including whether they will remain active donors or change their behavior to become mainly recipients, and vice-versa. These algorithms allowed us to analyze the factors that significantly influence behavior predictions. To evaluate our algorithms, we utilized data from over 2.4 million users on a peer-to-peer food-sharing online platform. Using our algorithm, we identified four main types of key user behavior patterns that occur over time. Moreover, we succeeded in forecasting future active donor key users and predicting the key users that would change their behavior to donors, with an accuracy of up to 89.6%. These findings provide valuable insights into the behavior of key users in volunteer-based social networks and pave the way for more effective communities-building in the future, while using the potential of machine learning for this goal.

Dockless e-scooters, a key micromobility service, have emerged as eco-friendly and flexible urban transport alternatives. These services improve first and last-mile connectivity, reduce congestion and emissions, and complement public transport for short-distance travel. However, effective management of these services depends on accurate demand prediction, which is crucial for optimal fleet distribution and infrastructure planning. While previous studies have focused on analyzing spatial or temporal factors in isolation, this study introduces a framework that integrates spatial, temporal, and network dependencies for improved micromobility demand forecasting. This integration enhances accuracy while providing deeper insights into urban micromobility usage patterns. Our framework improves demand prediction accuracy by 27 to 49

Large Language Models (LLMs) rapidly reshape modern life, advancing fields from healthcare to education and beyond. However, alongside their remarkable capabilities lies a significant threat: the susceptibility of these models to jailbreaking. The fundamental vulnerability of LLMs to jailbreak attacks stems from the very data they learn from. As long as this training data includes unfiltered, problematic, or 'dark' content, the models can inherently learn undesirable patterns or weaknesses that allow users to circumvent their intended safety controls. Our research identifies the growing threat posed by dark LLMs models deliberately designed without ethical guardrails or modified through jailbreak techniques. In our research, we uncovered a universal jailbreak attack that effectively compromises multiple state-of-the-art models, enabling them to answer almost any question and produce harmful outputs upon request. The main idea of our attack was published online over seven months ago. However, many of the tested LLMs were still vulnerable to this attack. Despite our responsible disclosure efforts, responses from major LLM providers were often inadequate, highlighting a concerning gap in industry practices regarding AI safety. As model training becomes more accessible and cheaper, and as open-source LLMs proliferate, the risk of widespread misuse escalates. Without decisive intervention, LLMs may continue democratizing access to dangerous knowledge, posing greater risks than anticipated.

BACKGROUND:Proteins act through physical interactions with other molecules to maintain organismal health. Protein-protein interaction (PPI) networks have proved to be a powerful framework for obtaining insight into protein functions, cellular organization, response to signals, and disease states. In multicellular organisms, protein content varies between tissues, influencing tissue morphology and function. Weighted PPI networks, reflecting the likelihood of interactions in specific tissues, offer insights into tissue-specific processes and disease mechanisms. We hypothesized that detecting anomalous nodes in these networks could reveal proteins with key tissue-specific functions. RESULTS:Here, we introduce Weighted Graph Anomalous Node Detection (WGAND), a novel machine-learning algorithm to identify anomalous nodes in weighted graphs. WGAND estimates expected edge weights and uses deviations to generate anomaly detection features, which are then used to score network nodes. We applied WGAND to weighted PPI networks of 17 human tissues. High-ranking anomalous nodes were enriched for proteins associated with tissue-specific diseases and tissue-specific biological processes, such as neuron signaling in the brain and spermatogenesis in the testis. WGAND outperformed other methods in terms of area under the ROC curve and precision at K, highlighting its effectiveness in uncovering biologically meaningful anomalies. CONCLUSIONS:Our findings demonstrate WGAND's potential as a powerful tool for detecting anomalous proteins with significant biological roles. By identifying proteins involved in critical tissue-specific processes and diseases, WGAND offers valuable insights for discovering novel biomarkers and therapeutic targets. Its versatile algorithm is suitable for any weighted graph and is broadly applicable across various fields. The WGAND algorithm is available as an open-source Python library at https://github.com/data4goodlab/wgand.

Voice data is increasingly being used in modern digital communications, yet there is still a lack of comprehensive tools for automated voice analysis and characterization. To this end, we developed the VANPY (Voice Analysis in Python) framework for automated pre-processing, feature extraction, and classification of voice data. The VANPY is an open-source end-to-end comprehensive framework that was developed for the purpose of speaker characterization from voice data. The framework is designed with extensibility in mind, allowing for easy integration of new components and adaptation to various voice analysis applications. It currently incorporates over fifteen voice analysis components - including music/speech separation, voice activity detection, speaker embedding, vocal feature extraction, and various classification models. Four of the VANPY's components were developed in-house and integrated into the framework to extend its speaker characterization capabilities: gender classification, emotion classification, age regression, and height regression. The models demonstrate robust performance across various datasets, although not surpassing state-of-the-art performance. As a proof of concept, we demonstrate the framework's ability to extract speaker characteristics on a use-case challenge of analyzing character voices from the movie "Pulp Fiction." The results illustrate the framework's capability to extract multiple speaker characteristics, including gender, age, height, emotion type, and emotion intensity measured across three dimensions: arousal, dominance, and valence.

Tracking lawmakers’ voting behavior is vital for government transparency. Although many parliaments publish voting records online, these datasets are often complex and difficult to analyze. This research presents the Voting Prediction Framework (VPF), a machine learning-based system designed to predict parliamentary voting outcomes at both the individual legislator and bill levels across multiple countries. VPF aims to support legislative transparency, streamline parliamentary work, and aid in policy refinement by identifying bills with a low likelihood of passing. The framework comprises three main components: (1) Data Collection – retrieving voting records from various countries using APIs, web crawlers, and structured databases; (2) Parsing and Feature Integration – enriching data with relevant features such as legislator seniority and bill content; and (3) Prediction Models – applying machine learning algorithms to forecast individual votes and bill-level outcomes. VPF is released as open source to encourage accessibility and adaptation. To validate VPF, we analyzed over 5 million votes from five countries: Canada, Israel, Tunisia, the United Kingdom, and the United States. Results indicate up to 85% precision in predicting individual votes and up to 84% accuracy for bill outcomes, demonstrating VPF’s effectiveness as a tool for political analysis and public insight into legislative processes.

Cuneiform tablets, emerging in ancient Mesopotamia around the late fourth millennium BCE, represent one of humanity's earliest writing systems. Characterized by wedge-shaped marks on clay tablets, these artifacts provided insight into Mesopotamian civilization across various domains. Traditionally, the analysis and dating of these tablets rely on subjective assessment of shape and writing style, leading to uncertainties in pinpointing their exact temporal origins. Recent advances in digitization have revolutionized the study of cuneiform by enhancing accessibility and analytical capabilities. Our research uniquely focuses on the silhouette of tablets as significant indicators of their historical periods, diverging from most studies that concentrate on textual content. Utilizing an unprecedented dataset of over 94,000 images from the Cuneiform Digital Library Initiative collection, we apply deep learning methods to classify cuneiform tablets, covering over 3,000 years of history. By leveraging statistical, computational techniques, and generative modeling through Variational Auto-Encoders (VAEs), we achieve substantial advancements in the automatic classification of these ancient documents, focusing on the tablets' silhouettes as key predictors. Our classification approach begins with a Decision Tree using height-to-width ratios and culminates with a ResNet50 model, achieving a 61 introduce novel VAE-powered tools to enhance explainability and enable researchers to explore changes in tablet shapes across different eras and genres. This research contributes to document analysis and diplomatics by demonstrating the value of large-scale data analysis combined with statistical methods. These insights offer valuable tools for historians and epigraphists, enriching our understanding of cuneiform tablets and the cultures that produced them.

The unprecedented growth in video conferencing usage is accompanied by multiple security and privacy threats. Importantly, protecting users' privacy is not always in their own hands. Posting meeting images affects all participants, leading to an easy collection of personal data including age, gender and linkage with participation in other meetings. Here, we explored privacy issues that may be at risk by attending virtual meetings. We extracted private information from collage images of meeting participants that are publicly posted online. We used image processing, text recognition tools, as well as social network analysis to explore our curated dataset of over 15 700 collage images, and over 142 000 face images of meeting participants. We demonstrate that video conference users are facing prevalent security and privacy threats. Our results indicate that it is relatively easy to collect thousands of publicly available images of video conference meetings and extract personal information about the participants, including their face images, age, gender, usernames, and even full names. This type of data can vastly and easily jeopardize people's security and privacy both in the online and real-world, affecting not only adults but also more vulnerable segments of society, such as children and older adults. Finally, we show that cross-referencing facial image data with social network data may put participants at additional privacy risks they may not be aware of and that it is possible to identify users that appear in several video conference meetings, thus providing a potential to maliciously aggregate different sources of information about a target individual.

In the last decades, global awareness toward the importance of diverse representation has been increasing. The lack of diversity and discrimination toward minorities did not skip the film industry. Here, we examine ethnic bias in the film industry through commercial posters, the industry’s primary advertisement medium for decades. Movie posters are designed to establish the viewer’s initial impression. We developed a novel approach for evaluating ethnic bias in the film industry by analyzing nearly 125,000 posters using state-of-the-art deep learning models. Our analysis shows that while ethnic biases still exist, there is a trend of reduction of bias, as seen by several parameters. Particularly in English-speaking movies, the ethnic distribution of characters on posters from the last couple of years is reaching numbers that are approaching the actual ethnic composition of the US population. An automatic approach to monitoring ethnic diversity in the film industry, potentially integrated with financial value, may be of significant use for producers and policymakers.

Embedding news articles is a crucial tool for multiple fields, such as media bias detection, identifying fake news, and making news recommendations. However, existing news embedding methods are not optimized to capture the latent context of news events. Most embedding methods rely on full-text information and neglect time-relevant embedding generation. In this paper, we propose a novel lightweight method that optimizes news embedding generation by focusing on entities and themes mentioned in articles and their historical connections to specific events. We suggest a method composed of three stages. First, we process and extract events, entities, and themes from the given news articles. Second, we generate periodic time embeddings for themes and entities by training time-separated GloVe models on current and historical data. Lastly, we concatenate the news embeddings generated by two distinct approaches: Smooth Inverse Frequency (SIF) for article-level vectors and Siamese Neural Networks for embeddings with nuanced event-related information. We leveraged over 850,000 news articles and 1,000,000 events from the GDELT project to test and evaluate our method. We conducted a comparative analysis of different news embedding generation methods for validation. Our experiments demonstrate that our approach can both improve and outperform state-of-the-art methods on shared event detection tasks.

Background Anomaly detection in graphs is critical in various domains, notably in medicine and biology, where anomalies often encapsulate pivotal information. Here, we focused on network analysis of molecular interactions between proteins, which is commonly used to study and infer the impact of proteins on health and disease. In such a network, an anomalous protein might indicate its impact on the organism’s health.Results We propose Weighted Graph Anomalous Node Detection (WGAND), a novel machine learning-based method for detecting anomalies in weighted graphs. WGAND is based on the observation that edge patterns of anomalous nodes tend to deviate significantly from expected patterns. We quantified these deviations to generate features, and utilized the resulting features to model the anomaly of nodes, resulting in node anomaly scores. We created four variants of the WGAND methods and compared them to two previously-published (baseline) methods. We evaluated WGAND on data of protein interactions in 17 human tissues, where anomalous nodes corresponded to proteins with major roles in tissue contexts. In 13 of the tissues, WGAND obtained higher AUC and P@K than baseline methods. We demonstrate that WGAND effectively identified proteins that participate in tissue-specific processes and diseases.Conclusion We present WGAND, a new approach to anomaly detection in weighted graphs. Our results underscore its capability to highlight critical proteins within protein-protein interaction networks. WGAND holds the promise to enhance our understanding of intricate biological processes and might pave the way for novel therapeutic strategies targeting tissue-specific diseases. Its versatility ensures its applicability across diverse weighted graphs, making it a robust tool for detecting anomalous nodes.### Competing Interest StatementThe authors have declared no competing interest.