Unverified50% confidenceFactExact time
Recognition of the second line of text used K-best dynamic programming, searching a dictionary of 17,700 words for the most likely combination.
1
Sources
50%
Confidence
Long-term
Relevance
10/15/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedK-best dynamic programming keeps the top K optimal candidate paths at each step, and is widely used in beam search decoding in speech recognition and machine translation.75% similarUnverifiedThe built-in dictionary can recognize common inflected forms of words (noun plurals, third-person singular verb forms, past tenses) and match them to their root forms.64% similarUnverifiedSemantic clustering is based on word embedding or sentence embedding technology, mapping keywords to points in a high-dimensional vector space and automatically grouping them using clustering algorithms such as K-means and DBSCAN.62% similarUnverified用户提问时问题被转换为向量，通过余弦相似度算法找到语义最相近的文档片段，这是语义检索优于传统关键词检索的根本原因62% similarVerifiedSemantic search converts text into high-dimensional vectors and finds the most relevant conclusions by computing cosine similarity, allowing queries phrased in different ways to match semantically equivalent results.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1039004API
curl https://kongchang.com/api/v1/knowledge/claims/1039004MCP
get_claim(id=1039004)