Media Cloud: Massive Open Source Collection of Global News on the Open Web
Abstract
We present the first full description of Media Cloud, an open source platform based on crawling hyperlink structure in operation for over 10 years, that for many uses will be the best way to collect data for studying the media ecosystem on the open web. We document the key choices behind what data Media Cloud collects and stores, how it processes and organizes these data, and its open API access as well as user-facing tools. We also highlight the strengths and limitations of the Media Cloud collection strategy compared to relevant alternatives. We give an overview two sample datasets generated using Media Cloud and discuss how researchers can use the platform to create their own datasets.
Keywords
Cite
@article{arxiv.2104.03702,
title = {Media Cloud: Massive Open Source Collection of Global News on the Open Web},
author = {Hal Roberts and Rahul Bhargava and Linas Valiukas and Dennis Jen and Momin M. Malik and Cindy Bishop and Emily Ndulue and Aashka Dave and Justin Clark and Bruce Etling and Rob Faris and Anushka Shah and Jasmin Rubinovitz and Alexis Hope and Catherine D'Ignazio and Fernando Bermejo and Yochai Benkler and Ethan Zuckerman},
journal= {arXiv preprint arXiv:2104.03702},
year = {2021}
}
Comments
15 pages, 9 figures, accepted (minus the 3-page, 3-image appendix given here) for publication and forthcoming in Proceedings of the Fifteenth International AAAI Conference on Web and Social Media (ICWSM-2021)