Model Predicted Data

Model Predicted Data

This folder contains our NLP and ML datasets as well as the dataset of accepted/rejected papers with stance values predicted by our model.

• 🗃 NLP dataset
• 🗃 ML dataset
• 🗃 Dataset of accepted/rejected papers

NLP and ML datasets

Each json file contains a list of papers with the following keys:

Key	Value	Description
url	string	url of the paper
title	string	title of the paper
author	string	author(s) of the paper
year	number	year the paper was published
venue	string	venue the paper was published in
citations	number	(optional) number of citations the paper has received, if known
stance	number	model predicted stance value between -1 (negative) and +1 (positive)

Dataset of accepted/rejected papers

The dataset contains papers compiled by us and papers compiled by Kang et al. (2018) (allenai/PeerRead).

Each json file contains a list of papers with the following keys:

Key	Value	Description
url	string	(optional) url of the paper, if known
title	string	title of the paper
author	string	author(s) of the paper
year	number	year the paper was published
venue	string	venue the paper was published in
status	string	"accepted" if the paper was accepted, "rejected" otherwise
stance	number	model predicted stance value between -1 (negative) and +1 (positive)