Detail 1
BERT language model(KorBERT) developed reflecting the characteristics of the Korean language is provided in open API services (http://aiopen.etri.re.kr) provided by ETRI
MoveDetail 1
Morphemes analysis, named entity recognition, and parsing have been selected as the Telecommunication Technology Association (TTA) standards through validation by various experts under direction of the ETRI language intelligence research section. The standards for thematic role recognition and question analysis are submitted and currently subject to validation. You can download the standards below
MoveDetail 1
Exobrain QA datasets (ETRI), language analysis corpora (ETRI), Korean TimeBank and SpaceBank, Morphology/Semantics corpora provided by University of Ulsan, corpora service in open API services (http://aiopen.etri.re.kr) provided by ETRI
MoveDetail 1
Journal of Society for Information Science and Technology, Vol. 35, No. 8, Aug. 2017
DownloadDetail 3(2017~2019)
Ulsan University corpus provides stemming and homonym annotation corpus (UTagger-HG), dependency and semantic annotation corpus (Utagger-DP/SR), and multilingual-level semantic dictionary (UPropBank)
MoveDetail 3(2017~2019)
Korlex1.0 database is under free distribution. Search service on web for Korlex1.5
MoveDetail 3(2017~2019)
Detail 3(2017~2019)