Source-linked AI summary
F -- A Model of Events based on the Foundational Ontology DOLCE+DnS Ultralite
Ansgar Scherp, Thomas Franz, Carsten Saathoff, Steffen Staab
TL;DR
Distributed event-based systems lack a formal event model that can support interoperability across heterogeneous components. The paper develops Event-Model-F on DOLCE+DnS Ultralite using modular, pattern-oriented ontologies, and reports comprehensive support for event structure and differing interpretations. Its scope follows a metaphysical distinction between events and objects that the authors note is not undisputed.
Problem
Existing event models inadequately support complex human-experience events and have ambiguous semantics, hindering interoperability among distributed event-based systems.
Method
Event-Model-F is a formal event model based on DOLCE+DnS Ultralite and developed with a pattern-oriented, modular ontology design.
Results
The analysis finds that existing models substantially lack structural relationships and alternative event interpretations, while Event-Model-F fulfills all stated non-functional requirements.
Takeaways & Limitations
Event-Model-F supports representing arbitrary real-world occurrences, event relations, and different interpretations while allowing modular management and domain-specific extension.
Takeaways & Limitations
The event–object distinction follows DOLCE+DnS Ultralite despite being metaphysically undisputed.
Abstract
from arXiv · showhide
The lack of a formal model of events hinders interoperability in distributed event-based systems. In this paper, we present a formal model of events, called Event-Model-F. The model is based on the foundational ontology DOLCE+DnS Ultralite (DUL) and provides comprehensive support to represent time and space, objects and persons, as well as mereological, causal, and correlative relationships between events. In addition, the Event-Model-F provides a flexible means for event composition, modeling event causality and event correlation, and representing different interpretations of the same event. The Event-Model-F is developed following the pattern-oriented approach of DUL, is modularized in different ontologies, and can be easily extended by domain specific ontologies.
1. INTRODUCTION
Distributed event-based systems increasingly combine heterogeneous components that detect, annotate, and exchange events, but existing models inadequately represent complex human experiences. Event-Model-F is introduced as a formal representation intended to support interoperability across these systems.
- Distributed event-based systems connect components that take events as input and provide events as output.Such systems arise as event detection and communication spread across sensors, software components, and proprietary solutions.
- Existing event models are conceptually narrow, lack systematic development, and have ambiguous semantics.
- Human-experience events require representation of time, space, participants, and mereological, causal, and correlative relationships.These events may also be subject to discussion and differing human interpretations.
- Event-Model-F provides a formal representation for capturing and representing human experience.The paper presents it as a model for events rather than only low-level technical signals.
- The model is designed to support event information interchange between event-based components and systems.Its scope includes causal relationships and different interpretations of the same event.
2. EVENTS AND OBJECTS
The paper distinguishes events from objects by following DOLCE+DnS Ultralite’s design decision. Events occur and unfold over time, whereas material objects exist and unfold over space.
- Events are treated as perduring entities that occur or happen and unfold over time.
- Material objects are treated as enduring entities that exist and unfold over space.
- Event-Model-F follows DOLCE+DnS Ultralite in distinguishing events from objects.The paper notes that this metaphysical distinction is not undisputed.
3. SCENARIO
The emergency-response scenario illustrates how multiple professional entities exchange descriptions of human-experience events across different event-based systems. A common formal representation is presented as useful for interoperability and machine-accessible semantics.
- The scenario involves an emergency hotline, police, fire department, control center, and forward liaison officers.
- A heavy storm and major flooding lead to events including a power outage and citizens reporting the outage to the emergency hotline.
- Hotline officers record calls and enter event descriptions that are annotated with call and recording information.
- The annotated event descriptions are automatically transferred to the emergency control center system.
- Different event-based systems need a common understanding of events to communicate efficiently across emergency-response entities.A formal representation supplies machine-accessible semantics for interoperability.
4. REQUIREMENTS ON F
The requirements combine functional coverage of event participation, time, space, structure, documentation, and interpretation with non-functional demands for extensibility, precision, modularity, reuse, and domain separation.
- Functional Requirements: Functional requirements were synthesized by analyzing event models from music, journalism, multimedia, news, cultural heritage, and knowledge representation.
- Functional Requirements: The model must represent participating living and non-living objects, including the roles they play in events.
- Functional Requirements: Events require temporal duration modeling, while objects require support for spatial extension and positioning.
- Functional Requirements: Structural support covers mereological, causal, and correlation relationships between events.The requirements include event composition, causes and effects, and correlation where causality may be unknown.
- Functional Requirements: The model must provide documentary support for events and objects through arbitrary annotations such as sensor and media data.
- Functional Requirements: Event interpretations must support different contextual points of view on the same real-world occurrences.
- Non-functional Requirements: Non-functional requirements include extensibility, axiomatization and formal precision, modularity, reuseability, and separation of concerns.These requirements support system evolution, machine checking, selective model use, reuse across domains, and integration of domain-specific knowledge.
5. DESIGN OF THE EVENT-MODEL-F
Event-Model-F aligns specialized event patterns with DOLCE+DnS Ultralite to represent contextualized events and their temporal, spatial, structural, documentary, and interpretive aspects.
- Foundational alignment: Event-Model-F aligns its functional requirements with DUL and uses the DnS pattern to represent contextualized views and interpretations of events.DnS reifies events and represents n-ary relations among event and object individuals.
- Participation and context: The participation pattern represents events, participating objects, their roles, and relevant spatial regions.Location parameters connect participant roles with space regions relevant to an event description.
- Event structure: The mereology pattern models composite events as wholes containing component events as parts.Component events can be constrained by temporal, spatial, and spatio-temporal conditions.
- Causality and correlation: The causality pattern classifies causes and effects and links their relationship to a justification representing an underlying theory.The justification may be an opinion, scientific law, or unspecified theory.
- Causality and correlation: The correlation pattern represents effects sharing a common cause, while the documentation pattern links events to documentary evidence such as sensor or media data.Correlation is explicitly modeled because the common cause may be unknown even when correlating effects are known.
- Interpretation: The interpretation pattern binds pattern instances into context-dependent views of the same event.Interpretations can include relevant situations satisfying participation, mereology, causality, correlation, and documentation relations.
6. USE OF EVENT-MODEL-F
The paper demonstrates Event-Model-F through an emergency-response scenario in which modular patterns describe and exchange a complex flood-related event across proprietary systems.
- Scenario modeling: The emergency scenario combines participation, mereology, causality, and correlation patterns to describe a flood and its related events.A flood can include multiple events, with causes and correlating events represented through separate patterns.
- Interpretation: Different causal opinions about the same flood are represented through multiple pattern instantiations organized by the interpretation pattern.Interpretation forms different nexuses of pattern instances and provides different points of view.
- Scenario modeling: A snapped power pole is modeled as the cause of a power outage, while houses and citizens participate in the outage event.The example defines snapped-power-pole-1 as Cause and power-outage-1 as Effect.
- Interoperability: Event-Model-F may help emergency-response entities integrate proprietary systems and communicate event descriptions.The use case requires entities using different data models to exchange event information.
7. EXISTING EVENT MODELS
The review finds that existing event models cover basic event aspects but provide limited support for structural relationships, multiple interpretations, formal semantics, and systematic modular development.
- Functional coverage: Existing models almost fully support participative, temporal, spatial, and documentary aspects of events.This assessment comes from the paper’s analysis of models across multiple domains.
- Functional coverage: Existing models substantially lack support for mereological, causal, and correlation relationships and different interpretations of the same event.Mereology is usually limited to simple part-of relations, causality lacks further axiomatization, and correlation appears only in a specific event-calculus extension.
- Interpretation support: None of the reviewed models provides different interpretations of the same event, while one treats this capability as future work.The comparison distinguishes this gap from models that support only technical event processing.
- Scope and approach: Related low-level event-processing systems focus on technical events within computerized systems rather than human-experience event interpretations.The paper contrasts publish/subscribe, complex event processing, and related systems with Event-Model-F’s scope.
- Scope and approach: Existing models generally lack systematic development, formal semantics, pattern-oriented structure, extensibility, and separation of concern.The paper states that Event-Model-F fulfills these non-functional requirements.
8. CONCLUSIONS
The paper concludes that Event-Model-F is a formal, modular event model aligned with DUL that supports structural relationships, multiple interpretations, and interoperability across event-based systems.
- Contributions: Event-Model-F uses DUL and pattern-oriented design to satisfy the paper’s non-functional requirements.Its patterns are specialized instantiations of the Descriptions and Situations model.
- Contributions: Its ontology patterns represent arbitrary real-world occurrences, structural event relations, and different interpretations of events.The supported structural aspect includes mereological, causal, and correlative relationships.
- Implications: Separating the model into smaller patterns helps manage event complexity and supports integration of event-based systems and components.The model’s formal nature enables this integration claim within the paper’s stated scope.
- Future work: Future work will investigate reasoning over pattern instantiations using metaknowledge and combining Event-Model-F with other core ontologies.The paper also identifies reusability as a future research direction.