dataset:
- name: LinkOrgs Training Data
tags:
- data-linkage
- record-linkage
- organizations
- LinkedIn-network
- bipartite-network
- Markov-network
configs:
- config_name: default
data_files:
- split: train
path: PosMatches_mat.parquet
license: mit
Introduction
Data repository for:
Brian Libgober, Connor T. Jerzak. Linking Datasets on Organizations Using Half-a-Billion Open-Collaborated Records. Political Science Methods and Research: 1-20, 2024. doi.org/10.1017/psrm.2024.55 arXiv
@article{libgober2024linking,
title={Linking Datasets on Organizations Using Half a Billion Open-Collaborated Records},
author={Libgober, Brian and Connor T. Jerzak},
journal={Political Science Methods and Research},
year={2024},
pages={1-20},
publisher={}
}
Details
This repository contains large-scale training data for improving linkage of data on organizations. NegMatches_mat.csv and NegMatches_mat_hold.csv refer to millions of negative name matches examples derived from the LinkedIn network (see paper for details). PosMatches_mat.csv and PosMatches_mat_hold.csv refer to millions of positive name matches examples derived from the LinkedIn network (see paper for details).
Additionally, files with saved *_bipartite refer to the bipartite network representation of the LinkedIn network that we use for improving linkage. files with saved *_bipartite refer to the Markov network representation of the LinkedIn network that we use for improving linkage.
Finally, data from all examples used in the paper are available in Example* folders. In each folder, the x and y data have linkage variables named by_x and by_y respectively, as does the merged z dataset.
Questions & Issues
Direct questions to: [email protected] or open an issue.
