Patent Search
The process of retrieving information from patent databases. Traditional approaches rely on keyword matching and classification code filtering, while emerging approaches introduce semantic search to improve recall and exploration efficiency.
Timeline (last 90 days)
Traditional patent searching mainly relies on keyword matching, meaning the system returns documents that contain the query terms.
In patent search scenarios, the cost of missed detection (low recall) is often higher than that of false detection (low precision), so researchers usually focus more on recall while controlling precision through ranking optimization.
Large language models can be used in patent search for query expansion, that is, automatically generating semantically equivalent or related search expressions based on the initial query, thereby improving recall.
All Facts (3)
Traditional patent searching mainly relies on keyword matching, meaning the system returns documents that contain the query terms.
50%UnverifiedIn patent search scenarios, the cost of missed detection (low recall) is often higher than that of false detection (low precision), so researchers usually focus more on recall while controlling precision through ranking optimization.
50%UnverifiedLarge language models can be used in patent search for query expansion, that is, automatically generating semantically equivalent or related search expressions based on the initial query, thereby improving recall.
50%