Dataset · clinc
Chatbot queries with out-of-scope examples
clinc/clinc_oos
An English intent classification corpus that teaches dialog systems to recognize irrelevant queries.
What's inside
Inside are English user queries annotated by supported intents, as well as examples of queries outside the bot's domain of competence (out-of-scope). The dataset size ranges from 10 to 100 thousand examples. The data is intended for solving the text classification task in target dialog systems.
What you can do with it
- Training an intent classifier for voice assistants and chatbots
- Testing models' ability to filter out queries outside supported topics
- Tasks
- Text classification
- Languages
- English
- Size
- 10K–100K rows
- Downloads this month
- 23.1K
- Likes
- 24
Models trained on it
- StrandsAgents/strands-decider-2B-qwen3.5-v1-2610Text classification 24.5K
- StrandsAgents/strands-decider-2B-hobson-v21Text classification 15.6K
- StrandsAgents/strands-decider-2B-hobson-v19Text classification 7K
- bartowski/RinggAI_ringg-router-e2b-GGUFMultimodal 3.7K
- StrandsAgents/strands-decider-E4B-gemma4-v1-2610Text classification 1.4K
- mradermacher/bobcat-flash-1.2-GGUF 1.1K
- StrandsAgents/strands-decider-12B-gemma4-v1-2610Text classification 1.1K
- manjunathshiva/opendecider-nanoZero shot classification 1K
How to load it
from datasets import load_dataset
ds = load_dataset("clinc/clinc_oos")