Automated Keyword Generation Using LLMs in a High-Volume Data Virtualization Service

The paper proposes a method for automatically generating keywords for virtual datasets in a data virtualization service (DVS) using large language models (LLMs). Five strategies for selecting text inputs to the LLMs were evaluated across five application case studies, with experiments involving three LLM generators and two embedding techniques. Results indicate that the effectiveness of each strategy depends on the quality of the text descriptions, leading to a multi‑strategy keyword generation service (KGS) integrated into the DVS.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

• llms
• rag
• computer-vision