CASino - Negotiation Dialogues in English
The CASino dataset contains simulated negotiation dialogues between two participants playing neighbours. Each exchange is annotated with negotiation strategies and enriched with metadata about the personality and satisfaction of the participants.
1,030 complete dialogues, JSON format with metadata on participants and strategies
CC-BY 4.0
Description
CASinO is a body of negotiation dialogues in English, where two participants negotiate resources in a fictional context of camping neighbors. The dataset includes strategic annotations for each line, as well as demographic and psychological metadata about the callers.
What is this dataset for?
- Train conversational models specialized in negotiation
- Analyze linguistic and behavioral strategies during negotiations
- Develop interactive conversational agents for collaborative or competitive purposes
Can it be enriched or improved?
Yes, it is possible to increase this dataset by adding more dialogs or by annotating the strategies used more finely. Collecting data in other languages or negotiation contexts would broaden its usefulness.
🔎 In summary
🧠 Recommended for
- Researchers in automated dialogue
- Conversational agent developers
- Pragmatic linguistics studies
🔧 Compatible tools
- Hugging Face Transformers
- Rasa
- By Lai
- Spacy
💡 Tip
Use strategic annotations to train models that can detect and adapt their negotiation tactics in real time.
Frequently Asked Questions
What is the size of the CASino dataset?
It includes 1,030 comprehensive negotiation dialogues between two participants.
Does this dataset contain annotations?
Yes, each dialogue is annotated with negotiation strategies and participant metadata.
Can this dataset be used for other languages or contexts?
For now, the dataset is only in English and in a camping context, but it can be extended to other contexts.
