Thesis Format
Monograph
Degree
Doctor of Philosophy
Program
Electrical and Computer Engineering
Collaborative Specialization
Artificial Intelligence
Supervisor
Kenneth McIsaac
2nd Supervisor
Abdulmutalib Wahaishi
Co-Supervisor
Abstract
In the realm of decentralized, epistemic environments, the interpretation of reality is inherently diverse and complex. This diversity stems from the multitude of autonomous entities, each with its unique set of beliefs and perspectives, leading to a wide array of ontological views. Such environments challenge the conventional ontology frameworks and their representation, which are predominantly designed from a singular, centralized observer’s viewpoint. These traditional approaches prove inadequate in addressing the nuances and multiplicity of interpretations that decentralized systems present.
This thesis explores the challenge of achieving semantic interoperability in decentralized systems, where diverse and autonomous systems operate without a unified framework. An Ontological View-Based Semantic Integration Framework (OVSIF) is introduced, aiming to enhance collaboration and information exchange across disparate systems through a novel approach to semantic integration.
Grounded in a thorough literature review, the work identifies significant gaps in existing semantic integration strategies. Building on this, a novel classification of environments— closed, decentralized, or open—is proposed. This classification aids in applying semantic strategies tailored to the specific challenges characteristic of each environment type. Following this, a new conceptualization structure is developed to align with the epistemic nature of decentralized systems, utilizing modal logic as the representation language to support precise semantic mappings. This setup promotes a hybrid approach that accommodates both intensional and extensional semantics.
The proposed framework for Semantic Integration in this research introduces an approach for integrating diverse ontological views without the necessity for a global consensus. This approach allows for adaptable and context-specific solutions. The practical applicability of the framework is demonstrated through logical analysis, theorem proving, and specific applications in the healthcare application domain, which illustrates its potential pertinence to improve operational efficiency and semantic coherence.
Summary for Lay Audience
Imagine a world where different computer systems, each with their own languages and rules, need to talk to each other effectively. This is a common scenario in areas like healthcare, where systems from different hospitals must communicate to provide the best care. However, this communication is often not straightforward due to the differences in how each system interprets information.
This is the essence of my research, in which a framework called the Ontological View- Based Semantic Integration (hence-after abbreviated as OVSIF) was developed to facilitate mutual understanding among these varied systems and thus enhance the efficiency of information sharing.
In this work, we explored how different environments (such as closed, open, or decentralized systems) affect the way information should be integrated. The proposed framework uses modal logic, a type of logic that deals with concepts like possibility and necessity. Modal logic helps to map out different ’possible worlds’—a way to represent each system’s own version of reality.
The framework can be utilized in many application domains. For example. In healthcare, for instance, OVSIF facilitates enhanced collaboration among doctors and clinicians specializing in diverse fields, particularly in addressing intricate cases like those involving patients with multiple conditions such as diabetes and kidney disease. Each physician offers insights into the patient’s health from their unique specialty, and thus OVSIF facilitates the integration of these perspectives without imposing a uniform approach. This ensures that each clinician’s viewpoint is acknowledged and leveraged effectively.
Although the work predominantly incorporates established tools such as modal logic, my approach in utilizing them to develop OVSIF is innovative and customized specifically to address the complexities of systems initially designed to function independently but now require collaboration.
The work adds to the advancement of distributed information systems, enhancing their intelligence and fostering cooperation, a critical aspect in addressing the intricate challenges of today’s interconnected world.
Recommended Citation
Adhnouss, Fateh Mohamed Ali, "Ontological View-based Semantic Integration Framework in Decentralized Environments" (2024). Electronic Thesis and Dissertation Repository. 10176.
https://ir.lib.uwo.ca/etd/10176
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.