The ATIS (Airline Travel Information System) Dataset includes spoken queries (i.e., utterances) annotated for the task of slot filling in conversational systems.
This dataset, ATISNER, includes airline spoken queries translated from English to Turkish, customized for Named Entity Recognition.
Train and test splits include 4,978 and 890 sentences, respectively.
Translations are provided by the following study.
Şahinuç, F., Yücesoy, V., & Koç, A. (2020). Intent Classification and Slot Filling… See the full description on the dataset page:
https://huggingface.co/datasets/ctoraman/atis-ner-turkish.
