SEACrowd/mtop_intent_classification

This dataset contains annotated utterances from 6 languages, including Thai, for semantic parsing. Queries corresponding to the chosen domains are crowdsourced. Two subsets are included in this dataset: 'domain' (eg. 'news', 'people', 'weather') and 'intent' (eg. 'GET_MESSAGE', 'STOP_MUSIC', 'END_CALL')