ccpemv2 / README.md
aloobun's picture
Update README.md
09bfb65
metadata
license: cc-by-4.0
language:
  - en
size_categories:
  - n<1K
pretty_name: movies dialog dataset
task_categories:
  - text-generation

A modified dataset of English dialogs between a user and an assistant discussing movie preferences in natural language. In the foundational dataset, it initially had 502 records.

Dataset Information

  • Name: ccpemv2.jsonl
  • Version: Version 0.2
  • Modifications Included: Transformed to ### User:... ### Assistant:... format
  • Language: English
  • License: Creative Commons Attribution 4.0

Original Dataset Citation

@inproceedings{radlinski-etal-2019-ccpe,
title = {Coached Conversational Preference Elicitation: A Case Study in Understanding Movie Preferences},
author = {Filip Radlinski and Krisztian Balog and Bill Byrne and Karthik Krishnamoorthi},
booktitle = {Proceedings of the Annual Meeting of the Special Interest Group on Discourse and Dialogue ({SIGDIAL})},
year = 2019
}