addyo07/

query-classification-dataset

This dataset provides ground-truth query classification samples across multiple languages (English, Devanagari Hindi, Code-switched Hinglish) for intent and memory scope routing in conversational AI and AI assistant engines.

	📦 Version History




	


	v2/memory_scope_golden_v1.json (4-Class Multi-Lingual Scope Taxonomy)

Total Verified Samples: 22,006 items… See the full description on the dataset page: https://huggingface.co/datasets/addyo07/query-classification-dataset.