Enhancing retrieval-augmented generation for Thai question answering through cosine similarity and interactive feedback | |
| Author | Nutdanai Sritunya |
| Call Number | AIT Thesis no.ISE-24-10 |
| Subject(s) | Natural language generation (Computer science) Generative artificial intelligence Information retrieval--Automation |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Mechatronics and Machine Intelligence |
| Publisher | Asian Institute of Technology |
| Abstract | With the growing demand for intelligent question-answering systems, Retrieval Augmented Generation (RAG) has emerged as a promising approach, but it faces challenges in ensuring coherence and user-alignment, particularly for low-resource languages like Thai. This research aimed to enhance the performance and user experience of RAG systems for Thai question-answering in the domain of environmental pollution. We proposed a novel approach incorporating a cosine similarity-based filtering mechanism to improve topical coherence among retrieved documents and an interactive feedback system for enabling user input during retrieval and generation. Through extensive experiments and user studies, our integrated system demonstrated significant improvements over the baseline RAG system, with domain experts consistently rating it higher for response quality, relevance, and coherence. Moreover, most participants across diverse backgrounds favored the interactive nature and responsive AI capabilities of the enhanced system. While promising, further advancements can be achieved through continuous knowledge base expansion, multimodal interactions, and real-world deployment for iterative improvements. |
| Year | 2024 |
| Type | Thesis |
| School | School of Engineering and Technology |
| Department | Department of Industrial Systems Engineering (DISE) |
| Academic Program/FoS | Mechatronics and Machine Intelligence (MMI) |
| Chairperson(s) | Mongkol Ekpanyapong; |
| Examination Committee(s) | Chaklam Slipasuwanchai;Ekbordin Winijkul; |
| Scholarship Donor(s) | Royal Thai Government Fellowship; |
| Degree | Thesis (M. Eng.) - Asian Institute of Technology, 2024 |