Update README.md
Browse files
README.md
CHANGED
|
@@ -6,4 +6,54 @@ language:
|
|
| 6 |
size_categories:
|
| 7 |
- 1K<n<10K
|
| 8 |
---
|
| 9 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
size_categories:
|
| 7 |
- 1K<n<10K
|
| 8 |
---
|
| 9 |
+
|
| 10 |
+
|
| 11 |
+
# Dataset: Core Intents
|
| 12 |
+
|
| 13 |
+
The **Core Intents** dataset contains feedback-related utterances classified into seven key categories. This dataset is designed to help train voice assistants in handling fundamental types of feedback and adjusting their behavior accordingly.
|
| 14 |
+
|
| 15 |
+
## Labels
|
| 16 |
+
|
| 17 |
+
The dataset includes the following labels:
|
| 18 |
+
- **ask_clarify**: Requests for clarification, where the user seeks clearer explanations.
|
| 19 |
+
- **ask_confirmation**: Requests for confirmation, where the user asks to validate a previous command or statement.
|
| 20 |
+
- **ask_repeat**: Requests for repetition, where the user asks for something to be repeated.
|
| 21 |
+
- **negative_feedback**: Negative responses or corrections, indicating dissatisfaction or disagreement.
|
| 22 |
+
- **positive_feedback**: Positive responses, expressing satisfaction or agreement.
|
| 23 |
+
- **neutral_feedback**: Neutral responses, where the user shows indifference or general acceptance.
|
| 24 |
+
- **stop**: Commands indicating the user wants to halt an action or process.
|
| 25 |
+
|
| 26 |
+
## Examples
|
| 27 |
+
|
| 28 |
+
Sample utterances for each label:
|
| 29 |
+
|
| 30 |
+
- **ask_clarify**:
|
| 31 |
+
- "your response was not clear"
|
| 32 |
+
- "your words were not so clear to me"
|
| 33 |
+
|
| 34 |
+
- **ask_confirmation**:
|
| 35 |
+
- "can you check and confirm my last command please"
|
| 36 |
+
- "can you check and confirm it"
|
| 37 |
+
|
| 38 |
+
- **ask_repeat**:
|
| 39 |
+
- "would you try again please"
|
| 40 |
+
- "would you try what you said again"
|
| 41 |
+
|
| 42 |
+
- **negative_feedback**:
|
| 43 |
+
- "command wrong"
|
| 44 |
+
- "dammit, I did not say it"
|
| 45 |
+
|
| 46 |
+
- **positive_feedback**:
|
| 47 |
+
- "thanks, that's amazing"
|
| 48 |
+
- "that is excellent, much appreciated"
|
| 49 |
+
|
| 50 |
+
- **neutral_feedback**:
|
| 51 |
+
- "anything is fine"
|
| 52 |
+
- "any one would be good to me"
|
| 53 |
+
|
| 54 |
+
## Purpose
|
| 55 |
+
|
| 56 |
+
This dataset aims to cover core interactions a voice assistant should manage to provide responsive and adaptive behavior in real-time communication. It can be used in tasks such as:
|
| 57 |
+
- Intent classification
|
| 58 |
+
- Dialogue system training
|
| 59 |
+
- Feedback management in voice assistants
|