ASC4AI: A Novel Automatic Service Composition Framework for AI Systems

Zhiguo Li, Qin Li*

*Corresponding author for this work

Research output: Contribution to journalConference articlepeer-review

1 Scopus citations

Abstract

As cloud computing becomes more prevalent in various domains, such as e-commerce, healthcare, education, etc., there is a growing demand for cloud-based applications that can perform complex tasks by integrating multiple cloud services. Despite QoS-aware automatic service composition has been extensively studied, conventional approaches face limitations in evaluating intelligent services and ensuring correctness and reliability for AI systems. To address these challenges, we propose five metrics for differentiating between intelligent and non-intelligent services and evaluating service composition solutions based on user-defined metrics constraints. We also explore ways to improve system correctness and reliability, and adapt these approaches to fit with AI systems. Building on these insights, we propose ASC4AI, a novel automatic service composition framework designed specifically for AI systems, which can automatically generate service composition solutions that meet user-defined functional and metrics constraints. Furthermore, we implement ASC4AI as a user-friendly tool that minimizes technical complexity for developers.

Original languageEnglish
Pages (from-to)197-202
Number of pages6
JournalProceedings of the International Conference on Software Engineering and Knowledge Engineering, SEKE
Volume2023-July
DOIs
StatePublished - 2023
Event35th International Conference on Software Engineering and Knowledge Engineering, SEKE 2023 - Hybrid, San Francisco, United States
Duration: 1 Jul 202310 Jul 2023

Keywords

  • Artificial intelligent systems
  • Automatic service composition
  • Service composition patterns
  • Service metrics

Fingerprint

Dive into the research topics of 'ASC4AI: A Novel Automatic Service Composition Framework for AI Systems'. Together they form a unique fingerprint.

Cite this