Event-driven time series forecasting constitutes a fundamental challenge across multiple scientific domains. Existing deep learning approaches rely on the assumption of historical pattern continuity and thus exhibit inadequate responsiveness to exogenous shocks, while mainstream text-enhanced methods encode documents as singular dense vectors, causing the compression of fine-grained semantic information such as entity stances, event types, and causal propagation relationships. To address these limitations, this study proposes a tripartite methodological framework building upon multimodal graph neural networks and the Temporal Fusion Transformer architecture. The first component employs large language models to perform hierarchical semantic extraction that decomposes unstructured text into four structured layers comprising entity stance quantification, event ontology mapping, inter-event causal chain reasoning, and aggregate sentiment indicators, thereby preserving prediction-relevant information that would otherwise be lost through flattened vectorization. The second component constructs heterogeneous information graphs containing multiple node types and relation types to explicitly model cumulative effects, propagation effects, and synergistic effects among events through relational graph convolutional networks with relation-aware attention mechanisms. The third component utilizes semantic analysis to achieve rapid environmental state identification, enabling dynamic recalibration of feature importance weights according to prevailing conditions and thereby addressing the non-stationarity inherent in event-driven prediction tasks. Experimental validation using crude oil price forecasting as a representative scenario demonstrates that the proposed framework achieves 17.7% improvement over baseline methods in overall prediction accuracy, with improvement reaching 34.7% during event-intensive periods, while ablation studies confirm independent contributions from each component. The proposed methodology possesses domain generality and can be transferred to other event-driven prediction tasks including epidemic spread forecasting, supply chain risk assessment, and financial market analysis.

Patient portal messaging has become a primary channel for asynchronous clinical communication, it spans a wide range of content, from symptom reports and medication concerns to administrative requests. Despite this volume and diversity, there is no formal representation for what a portal message contains: no vocabulary for the clinical and administrative events it describes, or for the attributes of those events that the patient has actually disclosed. Without such a representation, it is difficult to systematically analyze portal communication, assess message completeness, or build downstream tools that depend on structured input, such as automated triage, response drafting, and follow-up question generation. A clinical event schema, grounded in real portal messages and reviewed by clinicians, would provide this missing foundation. We introduce a clinical event ontology for patient portal messages, containing 8 event types and 70 roles that span clinical content (symptoms, medications, diagnostic tests, treatment responses, patient history) and administrative content (medical needs, logistics, social factors). The ontology was developed iteratively in collaboration with clinical expert and human evaluation. As a downstream application, we use the ontology to characterize the event types and roles most frequently sought in clinician follow-up questions, which provides insight of what clinicians ask about when reading portal messages.

EOAC-LLM: An LLM-driven event ontology automatic construction system

This research offers a fundamentally new perspective on the intellectual dynamics of 20th-century French philosophy through the lens of a non-canonical appropriation of Karl Marx's heritage. The novelty of the study lies in identifying a unified problematic field (from epistemology to ontology of events), formed by a radical rethinking of Marx's critique. For the first time in this context, the key role of Gilbert Simondon's thought is considered as a conceptual bridge between Althusser's structuralist "scientific" reading and Deleuze's post-structuralist philosophy of flows and becoming. In this examination, the hidden line from Simondon to Deleuze reveals an immanent transition from ideology critique to the ontology of productive forces, which constitutes the main internal tension of "French post-Marxism." The article demonstrates how central categories of contemporary thought were constructed through dialogue with Marx: the ideological state apparatus in Althusser, biopolitics in Foucault, control society in Deleuze, and event-based politics of truth in Badiou. The methodology of the study is built on a strategy of conceptual confrontation that reveals productive tensions between key readings of Marx. The rejection of a linear history of ideas allows tracking how conflicting ontological premises reinvent Marx's categories. Through a problem-oriented confrontation, the transformation of critique from ideology to the ontology of power and ethics of subjectivation is demonstrated. The result is a map of philosophical divergence that asserts the heuristic greatness of Marx as a source of radical questions of modernity. This approach consciously provokes a philosophical conflict of interpretations to reveal the radical heterogeneity within the very tradition of "French Marxism." The relevance of this intellectual reconstruction is determined by the contemporary crisis of legitimation, pervasive financial abstraction, and new forms of alienation. The philosophical tools forged by Althusser, Simondon, Foucault, Deleuze, and Badiou in working with Marx's heritage are now essential for diagnosing the society of digital platforms, algorithmic governance, and atomized individualism. This is why the unorthodox reception of Marx uncovers the critical potential of his thought that remains untapped within both traditional Marxism and the liberal consensus. They offer not ready-made answers but powerful methodologies for analyzing power, desire, and revolutionary subjectivation. Ultimately, Marx's greatness, re-embodied in French thought, appears not as an archival monument but as a living source for understanding and overcoming the abstract violence of capitalist modernity.

The Ontology of Adverse Events (OAE) was launched in 2011 to define, standardize and integrate various adverse events (AEs) arising after medical interventions. The terminological framework of OAE has undergone consistent expansion since its inception, driven by its successful implementation in numerous AE investigations. In this paper, we document substantial ontological extensions addressing patient anatomic regions and clinical manifestations, encompassing symptoms, physical signs, and pathological processes. Current statistical analysis reveals that OAE has 10,829 formally defined terms with unique identifiers. Compared to the 3,088 ontology terms included in the last OAE publication in 2014, 7,741 new terms have been added to OAE, which represents significant progress of the ontology in clinical granularity and domain coverage. The OAE framework enables structured representation of critical determinants influencing clinical outcomes, including but not limited to administration routes, dosage parameters, and demographic variables such as patient age. Through its standardized semantic architecture, OAE provides an integrative platform for cross-disciplinary analysis of AE patterns, etiological factors, and outcome trajectories in clinical interventions.

Adverse events (AEs) associated with COVID-19 vaccines remain a critical aspect of safety surveillance. In 2022, we reported the first systematic profiling of COVID-19 vaccine AEs using the Vaccine Adverse Event Reporting System (VAERS). Since then, vaccines have evolved with the introduction of bivalent formulations. This study provides an updated analysis to capture evolving safety trends. Building upon our previous analysis, we systematically analyzed AE profiles for the Pfizer-BioNTech, Moderna, and Janssen vaccines, along with the newer bivalent Pfizer-BioNTech and Moderna vaccines and the protein subunit Novavax vaccine, using VAERS data through 28 June 2024. We obtained processed VAERS data via Cov19VaxKB. Significance of each AE was determined using Pearson's Chi-square test, proportional reporting ratios, and case report frequencies with established thresholds. Overlap and age- or sex-stratified analyses were conducted to characterize shared and unique AE patterns across vaccine types. AE classification using the Ontology of Adverse Events was performed to categorize and interpret significant AEs within a structured hierarchy. We observed a marked decrease in unique AEs reported for the Pfizer-BioNTech and Moderna monovalent mRNA vaccines and the recombinant vector vaccine Janssen. The bivalent versions of Pfizer-BioNTech and Moderna exhibited distinct overlapping AE profiles compared to their monovalent counterparts, and bivalent vaccines were generally associated with fewer AEs than the classical monovalent vaccines. Significant differences were observed in thrombosis, myocarditis, and Guillain-Barré syndrome (GBS) across vaccines. Age-specific analyses revealed a bimodal pattern with higher AE reporting in children aged 0-9 and adults aged 50-69, and clear sex differences. Females reported more common AEs, while males were more often linked to serious AEs (thrombosis, myocarditis, and GBS). Death-related AEs were uncommon but more frequent among older males and primarily associated with monovalent formulations. Ontology-based classification revealed that females were more likely to experience sensory-related AEs, whereas males were more prone to cardiovascular-related AEs. The adverse event profiles of COVID-19 vaccines during 2020-2024 largely overlapped those identified during 2020-2021, while also revealing new overall and age- and sex-specific AE patterns. Ontology-guided classification enhanced the interpretation of large-scale vaccine safety data, supporting more precise risk assessment across groups.

This study is a detailed analysis of the influence of the philosophy of G.W. Leibniz on the metaphysics and ethics of Gilles Deleuze, with a focus on the transformation of Leibniz’s key concepts within Deleuze’s philosophical system. Particular attention is paid to the evolution of Deleuze’s perception of Leibniz’s philosophy: from the critical perception presented in early works such as “Difference and Repetition” and “Expressionism in Philosophy: Spinoza”, where Leibniz is interpreted as a philosopher of transcendental representation, to the integration of Leibniz’s key concepts into Deleuze’s own philosophical system, which is particularly evident in “The Fold. Leibniz and the Baroque”. The study focuses on Deleuze’s revision of the principles of identity and non-contradiction - which he reinterprets in the context of his ontology of events, series and singularities. The focus is on how Deleuze replaces Leibniz’s theological perspective with an immanent system in which the harmony of the world is based on a mathematically optimal distribution of singularities. Deleuze’s ethical project is examined through the lens of the concept of puissance (force, power, potency), which combines the ideas of Nietzsche, Spinoza and Leibniz. The study emphasises the significance of interrelated categories such as the will to power, the capacity for action and potency, which become the basis of Deleuze’s ethical system. Chaos and difference are seen not as destructive forces, but as grounds for the formation of a dynamic harmony that unites disparate elements into a coherent system. The study is characterized by a unique approach that combines Deleuze’s metaphysics and ethics, as well as an original interpretation of his interaction with Leibniz’s philosophy. The author argues that Deleuze forms a unique view of the world as a process where chaos and difference act as the foundations for dynamic harmony. This approach allows Deleuze to combine elements of Leibniz’s and Nietzsche’s theories, presenting our world as the best possible world because of its ability to maintain maximum variety while preserving wholeness.

Abstract This paper revisits Pietroski’s challenge to truth-conditional semantics, focusing on his critique of Davidsonian event semantics in the case of mutual chases , which provide an intriguing exemplification of the complex interaction between logical forms and ambiguous contexts of utterance. Pietroski argues that truth-conditional analyses either collapse into contradiction or demand dubious metaphysical commitments, pushing us instead toward a purely procedural, concept-assembly view of meaning. Building on Quine’s “double-vision” puzzle, we defend truth-conditional semantics by introducing the notion of event guises : role-indexed properties of a single complex event. We argue that guises are supported not by subevents or perspectival constructs, but by minimal subsituations in Kratzer’s sense, i.e. informational parts of the situation that supports the event. This analysis dissolves apparent contradictions, preserves a lean ontology of events, and accommodates adverbial modification without metaphysical extravagance. The upshot is a hybrid event–situation semantics that shows why logical forms remain robustly truth-conditional, contra Pietroski’s skepticism.

Abstract. This study pioneers an event ontology framework for Chinese Ancient Villages with Red Legacy to resolve cultural resource fragmentation and narrative discontinuity. Extending the CIDOC-CRM standard, this paper developed a multidimensional ontology integrating villages, events, figures, locations, time, and cultural resources through domain-specific semantic modeling. Utilizing Protégé and Neo4j tools, unstructured cultural resources from 754 nationally designated villages were transformed into structured linked data, exemplified by the Dawangmiao Village knowledge graph implementation. This framework enables multidimensional access to tangible/intangible heritage, dynamic reconstruction of revolutionary history via semantic relationships, and structural representation of spatiotemporal contexts. The ontology-driven approach significantly enhances scholarly research capabilities and dissemination efficacy, establishing a reusable digital infrastructure for transmedia storytelling. It provides methodological foundations for digital conservation of red legacy, directly supporting red-themed education and sustainable cultural tourism development through semantically enriched knowledge organization.

Event relation extraction plays a crucial role in constructing an event knowledge graph. However, current models only extract trigger words as event ontology representations, and do not consider node type during information aggregation, resulting in low accuracy in event relation extraction. To address these challenges, we propose an event relation extraction model based on heterogeneous graph attention networks and event ontology direction induction. To enhance the completeness of event information, we incorporate argument role information, in addition to trigger words, into the input text. A novel heterogeneous graph attention framework is proposed to reasonably allocate weights to trigger words, argument roles, and text information, and then perform two levels of aggregation, node-level and semantic-level, in sequence. To improve the accuracy of event direction discrimination, we construct an event ontology subgraph that includes trigger words and arguments to aggregate complete event structure information during direction induction. Finally, we evaluate our model on three datasets, TimeBank-Dense, MATRES, and HiEve, and demonstrate that our model outperforms state-of-the-art models by 1.2%, 0.5%, and 0.8%, respectively, in terms of the Micro-F1 score. Our proposed model provides a promising solution for event relation extraction and can be applied in various natural language processing applications.

BackgroundVaccines are crucial for preventing infectious diseases; however, they may also be associated with adverse events (AEs). Conventional analysis of vaccine AEs relies on manual review and assignment of AEs to terms in terminology or ontology, which is a time-consuming process and constrained in scope. This study explores the potential of using Large Language Models (LLMs) and LLM text embeddings for efficient and comprehensive vaccine AE analysis.ResultsWe used Llama-3 LLM to extract AE information from FDA-approved vaccine package inserts for 111 licensed vaccines, including 15 influenza vaccines. Text embeddings were then generated for each vaccine’s AEs using the nomic-embed-text and mxbai-embed-large models. Llama-3 achieved over 80% accuracy in extracting AE text from vaccine package inserts. To further evaluate the performance of text embedding, the vaccines were clustered using two clustering methods: (1) LLM text embedding-based clustering and (2) ontology-based semantic similarity analysis. The ontology-based method mapped AEs to the Human Phenotype Ontology (HPO) and Ontology of Adverse Events (OAE), with semantic similarity analyzed using Lin’s method. Text embeddings were generated for each vaccine’s AE description using the LLM nomic-embed-text and mxbai-embed-large models. Compared to the semantic similarity analysis, the LLM approach was able to capture more differential AE profiles. Furthermore, LLM-derived text embeddings were used to develop a Lasso logistic regression model to predict whether a vaccine is “Live” or “Non-Live”. The term “Non-Live” refers to all vaccines that do not contain live organisms, including inactivated and mRNA vaccines. A comparative analysis showed that, despite similar clustering patterns, the nomic-embed-text model outperformed the other. It achieved 80.00% sensitivity, 83.06% specificity, and 81.89% accuracy in a 10-fold cross-validation. Many AE patterns, with examples demonstrated, were identified from our analysis with AE LLM embeddings.ConclusionThis study demonstrates the effectiveness of LLMs for automated AE extraction and analysis, and LLM text embeddings capture latent information about AEs, enabling more comprehensive knowledge discovery. Our findings suggest that LLMs demonstrate substantial potential for improving vaccine safety and public health research.

BackgroundOver the last 10-15 years, US health care and the practice of medicine itself have been transformed by a proliferation of digital medicine and digital therapeutic products (collectively, digital health tools [DHTs]). While a number of DHT classifications have been proposed to help organize these tools for discovery, retrieval, and comparison by health care organizations seeking to potentially implement them, none have specifically addressed that organizations considering their implementation approach the DHT discovery process with one or more specific outcomes in mind. An outcomes-based DHT ontology could therefore be valuable not only for health systems seeking to evaluate tools that influence certain outcomes, but also for regulators and vendors seeking to ascertain potential substantial equivalence to predicate devices.ObjectiveThis study aimed to develop, with inputs from industry, health care providers, payers, regulatory bodies, and patients through the Accelerated Digital Clinical Ecosystem (ADviCE) consortium, an ontology specific to DHT outcomes, the Digital medicine Outcomes Value Set (DOVeS), and to make this ontology publicly available and free to use.MethodsFrom a starting point of a 4-generation–deep hierarchical taxonomy developed by ADviCE, we developed DOVeS using the Web Ontology Language through the open-source ontology editor Protégé, and data from 185 vendors who had submitted structured product information to ADviCE. We used a custom, decentralized, collaborative ontology engineering methodology, and were guided by Open Biological and Biomedical Ontologies (OBO) Foundry principles. We incorporated the Mondo Disease Ontology (MONDO) and the Ontology of Adverse Events. After development, DOVeS was field-tested between December 2022 and May 2023 with 40 additional independent vendors previously unfamiliar with ADviCE or DOVeS. As a proof of concept, we subsequently developed a prototype DHT Application Finder leveraging DOVeS to enable a user to query for DHT products based on specific outcomes of interest.ResultsIn its current state, DOVeS contains 42,320 and 9481 native axioms and distinct classes, respectively. These numbers are enhanced when taking into account the axioms and classes contributed by MONDO and the Ontology of Adverse Events.ConclusionsDOVeS is publicly available on BioPortal and GitHub, and has a Creative Commons license CC-BY-SA that is intended to encourage stakeholders to modify, adapt, build upon, and distribute it. While no ontology is complete, DOVeS will benefit from a strong and engaged user base to help it grow and evolve in a way that best serves DHT stakeholders and the patients they serve.

This article proposes a link between temporal structuring and the dynamics of organizing that is manifest in a fabric of concurrent paths that we call a path net. Path nets are shaped by mechanisms of temporal structuring, such as entrainment, planning, agency, and chance. Path nets materialize the effects of temporal structuring “here” and “now” in the comings and goings of actors and resources, thereby setting the stage for doings and sayings of situated practice and shaping the dynamics of organizing. Path nets offer a parsimonious system of picturing the complexity of organizing that is built on a processual ontology of paths and events. Through the lens of the path net, we can picture temporal structuring as a motor for organizing that drives recurrence without assuming stability or change. Comings and goings are readily observable, so path nets open new directions for empirical research on temporality and the dynamics of organizing.

This paper explores event ontology, a foundational philosophy of the materialist worldview, and presents an analysis of Nietzsche’s philosophical materialism, drawing upon his late notebooks, particularly his project on the Will to Power. Our approach situates Nietzsche’s perspective within the metaphysical direction of immanent materialism, and we draw connections between his ideas and the materialist monism of Russell, the process ontology of Whitehead, and the ontology of individuation developed by Simondon. The paper contributes to ongoing discussions on materialism in philosophy and sheds light on the intersection of Nietzsche’s thought with other prominent materialist theories.

Forecasting violence has become a critical obstacle in the field of video monitoring to guarantee public safety. Lately, YOLO (You Only Look Once) has become a popular and effective method for detecting weapons. However, identifying and forecasting violence remains a challenging endeavor. Additionally, the classification results had to be enhanced with semantic information. This study suggests a method for forecasting violent incidents by utilizing Yolov9 and ontology. The authors employed Yolov9 to identify and categorize weapons and individuals carrying them. Ontology is utilized for semantic prediction to assist in predicting violence. Semantic prediction happens through the application of a SPARQL query to the identified frame label. The authors developed a Threat Events Ontology (TEO) to gain semantic significance. The system was tested with a fresh dataset obtained from a variety of security cameras and websites. The VP Dataset comprises 8739 images categorized into 9 classes. The authors examined the outcomes of using Yolov9 in conjunction with ontology in comparison to using Yolov9 alone. The findings show that by combining Yolov9 with ontology, the violence prediction system's semantics and dependability are enhanced. The suggested system achieved a mean Average Precision (mAP) of 83.7 %, 88% for precision, and 76.4% for recall. However, the mAP of Yolov9 without TEO ontology achieved a score of 80.4%. It suggests that this method has a lot of potential for enhancing public safety. The authors finished all training and testing processes on Google Colab's GPU. That reduced the average duration by approximately 90.9%. The result of this work is a next level of object detectors that utilize ontology to improve the semantic significance for real-time end-to-end object detection.

Event extraction is a complex and challenging task in the field of information extraction. It aims to identify event types, triggers, and argument information from the text. In recent years, overlapping event extraction has attracted the attention of researchers because of its higher challenge and practicability, and some work has carried out in-depth research on overlapping event extraction and achieved remarkable results. But these works (1) ignore the role of ontology knowledge in event extraction; (2) use the same semantic encoding for multi-stage models, lacking consideration for the independent characteristics of extraction tasks such as event types, triggers, and arguments; and (3) face issues in the training process of multi-stage models, such as error cascading and slow convergence. To address the above issues, we propose an ontology-guided and scheduled-sampling approach for overlapping event extraction, termed as OGSS. First, we design a symmetric matrix for event ontology knowledge representation and integrate it into the semantic encoding process, infusing ontology knowledge into event extraction. Second, for extraction targets such as event types, triggers, and arguments, we process the semantic encoding according to the characteristics of each extraction target, obtaining semantic representations tailored for each subtask. Finally, we view multi-stage predictions as sequential outputs of a joint model, using a scheduled sampling strategy between subtasks to effectively mitigate the cascading propagation of errors during training and accelerate model convergence. We conduct extensive experiments on the FewFc event extraction benchmark dataset. The results show that OGSS achieves significant improvements in overlapping event extraction tasks compared to previous methods.

COfEE: A comprehensive ontology for event extraction from text

BackgroundIntraoperative neurophysiological monitoring (IOM) plays a pivotal role in enhancing patient safety during neurosurgical procedures. This vital technique involves the continuous measurement of evoked potentials to provide early warnings and ensure the preservation of critical neural structures. One of the primary challenges has been the effective documentation of IOM events with semantically enriched characterizations. This study aimed to address this challenge by developing an ontology-based tool.MethodsWe structured the development of the IOM Documentation Ontology (IOMDO) and the associated tool into three distinct phases. The initial phase focused on the ontology’s creation, drawing from the OBO (Open Biological and Biomedical Ontology) principles. The subsequent phase involved agile software development, a flexible approach to encapsulate the diverse requirements and swiftly produce a prototype. The last phase entailed practical evaluation within real-world documentation settings. This crucial stage enabled us to gather firsthand insights, assessing the tool’s functionality and efficacy. The observations made during this phase formed the basis for essential adjustments to ensure the tool’s productive utilization.ResultsThe core entities of the ontology revolve around central aspects of IOM, including measurements characterized by timestamp, type, values, and location. Concepts and terms of several ontologies were integrated into IOMDO, e.g., the Foundation Model of Anatomy (FMA), the Human Phenotype Ontology (HPO) and the ontology for surgical process models (OntoSPM) related to general surgical terms. The software tool developed for extending the ontology and the associated knowledge base was built with JavaFX for the user-friendly frontend and Apache Jena for the robust backend. The tool’s evaluation involved test users who unanimously found the interface accessible and usable, even for those without extensive technical expertise.ConclusionsThrough the establishment of a structured and standardized framework for characterizing IOM events, our ontology-based tool holds the potential to enhance the quality of documentation, benefiting patient care by improving the foundation for informed decision-making. Furthermore, researchers can leverage the semantically enriched data to identify trends, patterns, and areas for surgical practice enhancement. To optimize documentation through ontology-based approaches, it’s crucial to address potential modeling issues that are associated with the Ontology of Adverse Events.

We argue that a phenomenological analysis of consciousness similar to that of Husserl shows that the effects of phenomenal qualities shape our perception of the world. It also shows the way the physical and mathematical sciences operate, allowing us to accurately describe the observed regularities in terms of communicable mathematical laws. The latter say nothing about the intrinsic features of things. They only refer to the observed regularities in their behaviors, providing rigorous descriptions of how the universe works, to which any viable ontology must conform. Classical mechanistic determinism limits everything that can occur to what happens in an instant and leaves no room for novelty or any intrinsic aspect that is not epiphenomenal. The situation changes with quantum probabilistic determinism if one takes seriously the ontology that arises from its axioms of objects, systems in certain states, and the events they produce in other objects. As Bertrand Russell pointed out almost a century ago, an ontology of events with an internal phenomenal aspect, now known as panprotopsychism, is better suited to explaining the phenomenal aspects of consciousness. The central observation of this paper is that many objections to panpsychism and panprotopsychism, which are usually called the combination problem, arise from implicit hypotheses based on classical physics about supervenience. These are inappropriate at the quantum level, where an exponential number of emergent properties and states arise. The analysis imposes conditions on the possible implementations of quantum cognition mechanisms in the brain.

Abstract How do international crises unfold? We conceptualize international relations as a strategic chess game between adversaries and develop a systematic way to measure pieces, moves, and gambits accurately and consistently over a hundred years of history. We introduce a new ontology and dataset of international events called ICBe based on a very high-quality corpus of narratives from the International Crisis Behavior (ICB) Project. We demonstrate that ICBe has higher coverage, recall, and precision than existing state of the art datasets and conduct two detailed case studies of the Cuban Missile Crisis (1962) and the Crimea-Donbas Crisis (2014). We further introduce two new event visualizations (event iconography and crisis maps), an automated benchmark for measuring event recall using natural language processing (synthetic narratives), and an ontology reconstruction task for objectively measuring event precision. We make the data, supplementary appendix, replication material, and visualizations of every historical episode available at a companion website crisisevents.org.