Dataset · PolyAI
Banking queries with 77 intent labels
PolyAI/banking77
13,083 English customer banking queries labeled with 77 fine-grained intents for classification.
What's inside
A collection of online customer banking queries annotated with their corresponding intents, totaling 13,083 queries and 77 highly detailed classes within a single banking domain. The language is English, and the data is designed for intent detection.
What you can do with it
- Train an intent classifier for a banking chatbot
- Test text-classification models across 77 classes
- Fine-tune LoRA or encoders for granular intent detection
- Tasks
- Text classification
- Languages
- English
- Size
- 10K–100K rows
- Downloads this month
- 36.7K
- Likes
- 139
Models trained on it
- sileod/deberta-v3-base-tasksource-nliZero shot classification 8K
- mradermacher/bobcat-flash-1.2-GGUF 1.1K
- mradermacher/nexus-decision-4b-GGUF 1K
- mradermacher/elythros-decider-1.7b-GGUFZero shot classification 887
- mradermacher/elf-1-12b-GGUF 753
- mradermacher/elf-1-31b-GGUF 721
- mradermacher/bobcat-1.2-GGUF 715
- Shroomy/tinydistill-banking77-minilmText classification 648
How to load it
from datasets import load_dataset
ds = load_dataset("PolyAI/banking77")