Assembling Fragmented Domain Knowledge: A LLM-Powered QA System for Taiwan Cinema

En Chun Kuo, Yea Huey Su

研究成果: 書貢獻/報告類型會議論文篇章同行評審

摘要

This research introduces the development of a specialized Question Answering (QA) system, designed to tackle the challenges posed by dispersed domain knowledge. Specifically tailored for the Taiwanese movie industry, this system utilizes advancements in Natural Language Processing (NLP) and incorporates large language models via open-source platforms like LangChain. Our aim is to facilitate industry professionals in swiftly locating and extracting pertinent information from extensive data resources. A key focus is on mitigating the risk of data leakage, which is often associated with uploading documents to general-purpose chatbots. We have conducted a comprehensive evaluation of our Large Language Model (LLM)-powered QA system, showcasing its efficacy and accuracy in response. Ultimately, this research strives to illuminate the complexities of aggregating scattered expertise, aiding those who seek to delve deeply into domain-specific knowledge.

原文???core.languages.en_GB???
主出版物標題2024 IEEE Congress on Evolutionary Computation, CEC 2024 - Proceedings
發行者Institute of Electrical and Electronics Engineers Inc.
ISBN(電子)9798350308365
DOIs
出版狀態已出版 - 2024
事件13th IEEE Congress on Evolutionary Computation, CEC 2024 - Yokohama, Japan
持續時間: 30 6月 20245 7月 2024

出版系列

名字2024 IEEE Congress on Evolutionary Computation, CEC 2024 - Proceedings

???event.eventtypes.event.conference???

???event.eventtypes.event.conference???13th IEEE Congress on Evolutionary Computation, CEC 2024
國家/地區Japan
城市Yokohama
期間30/06/245/07/24

指紋

深入研究「Assembling Fragmented Domain Knowledge: A LLM-Powered QA System for Taiwan Cinema」主題。共同形成了獨特的指紋。

引用此