A manually annotated Latvian FrameNet annotation layer of the FullStack corpus.
Fields: [ID, FORM, LEMMA, UPOS, XPOS, FEATS, HEAD, DEPREL, DEPS] + [FILLPRED, PRED, APRED]
- FILLPRED: indicates the target word
- PRED: semantic frame
- APRED: frame elements
Note that each corpus example contains one FrameNet annotation set (frame instance). If there are several annotation sets for the same sentence, they are included as separate corpus examples.
- Annotation sets: 10,952
- Semantic frames: 458
- Lexical units: 1,658
- Target words (lexemes): 863