AskFlow: A Multi-Agent Framework for Intelligent Product Customer Support

Wang, Shudong (2025) AskFlow: A Multi-Agent Framework for Intelligent Product Customer Support. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (2MB) | Preview
Preview	PDF (Configuration Manual) Download (733kB) | Preview

Abstract

This thesis presents the design, implementation, and evaluation of AskFlow, a multi-agent framework for intelligent customer support with demonstrated effectiveness in educational domains and extensibility to healthcare applications. The system addresses key challenges in customer service automation by integrating multiple AI providers and implementing domain-specific document processing techniques. Through detailed case studies in educational institutions and development of healthcare evaluation frameworks, this research shows measurable improvements over traditional keyword-based support systems. The modular architecture supports multiple AI providers (OpenAI, Anthropic, Google Gemini, Cohere) and various document processing strategies, offering organizations a practical solution for domain-specific customer support automation.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Vamadevan, Arundev UNSPECIFIED
Uncontrolled Keywords:	Multi-Agent Systems; Customer Support Automation; Retrieval-Augmented Generation; Multi-Provider AI; Intelligent Information Systems
Subjects:	Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence H Social Sciences > HF Commerce > Customer Service
Divisions:	School of Computing > Master of Science in Artificial Intelligence
Depositing User:	Ciara O'Brien
Date Deposited:	12 Aug 2026 10:13
Last Modified:	12 Aug 2026 10:13
URI:	https://norma.ncirl.ie/id/eprint/9517

Actions (login required)

View Item