University of the Aegean
Semantic service discovery supported by Large Language Models
Abstract
dc:descriptionThis doctoral dissertation focuses on revitalizing the vision of Service-Oriented Architecture (SOA) through the use of Large Language Models (LLMs). Despite the initial success of SOA, the implementation of universal service registries failed due to the "broken triangle" (the failure of standards like UDDI), inaccurate service discovery, and the neglect of non-functional characteristics. The research concentrates on RESTful services and aims to overcome the limitations of traditional Information Retrieval and Machine Learning techniques, which lack semantic depth or suffer from a lack of training data.Key ContributionsFunctional Aspect: A method was developed for the automatic generation of rich OpenAPI specifications from source code using LLMs.An innovative semantic annotation method was introduced using cross-domain ontologies, achieving exceptional accuracy. Two advanced matching algorithms were proposed: one combining strict and "relaxed" semantic matching (Action-Input-Output) and a second based entirely on LLM prompting, which provides justification for results and nearly ideal accuracy. Non-Functional Aspect: The thesis introduces a mechanism for assessing service maturity (based on the Richardson maturity model). Multiple LLMs are employed to identify quality issues and eliminate "hallucinations," creating a comprehensive non-functional profile. Conclusion Adopting the Design Science Research methodology, the work concludes with the creation of a prototype semantic registry. By combining automated specification generation with high matching accuracy, the dissertation demonstrates that Artificial Intelligence can make global service discovery practically feasible, bridging the gap between academic theory and industrial application.
Degree
thesis:*- Grantor dc:publisher
- University of the Aegean
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Smardas, Antonios
- Σμαρδάς, Αντώνιος
Subjects
dc:subject × 8- Υπηρεσίες
- LLM
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Επιστήμες Μηχανικού και Τεχνολογία
- Μηχανική και συστήματα επικοινωνιών, Τηλεπικοινωνίες
- Electrical Engineering, Electronic Engineering, Information Engineering
- Engineering and Technology
- Communication engineering and systems, Telecommunications
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/61317