Skip to main content

Search tokens in dataset

GET 
/api/datasets/:slug/tokens

Search through tokens with mixed token/tag sequence matching.

Search Syntax​

The search parameter supports mixed token and tag conditions:

  • Token search: just the text (e.g., đi)
  • Tag search: prefix with tag: (e.g., tag:V)
  • Combined sequence: comma-separated (e.g., đi,tag:R finds "đi" followed by any adverb)

Examples​

  • search=đi - Find sentences containing the token "đi"
  • search=tag:N - Find sentences containing any noun
  • search=đi,tag:R - Find "đi" followed by an adverb (like "đi ra", "đi về")
  • search=tag:N,tag:V - Find noun followed by verb sequences

Token data is stored in R2 and parsed from VLSP format files.

Request​

Responses​

Token search results with sentence context