Determining the accuracy of crowdsourced tweet verification for auroral research

Case, Nathan Anthony and MacDonald, Elizabeth A. and McCloat, Sean and Lalone, Nicolas and Tapia, Andrea (2016) Determining the accuracy of crowdsourced tweet verification for auroral research. Citizen Science: Theory and Practice, 2016. ISSN 2057-4991

[thumbnail of Verified Tweet Validation]
Preview
PDF (Verified Tweet Validation)
Verified_Tweet_Validation.pdf - Accepted Version

Download (511kB)

Abstract

The Aurorasaurus citizen science project harnesses volunteer crowdsourcing to identify sightings of an aurora (or the "northern/southern lights") posted by citizen scientists on Twitter. Previous studies have demonstrated that aurora sightings can be mined from Twitter but with the caveat that there is a high level of accompanying non-sighting tweets, especially during periods of low auroral activity. Aurorasaurus attempts to mitigate this, and thus increase the quality of its Twitter sighting data, by utilizing volunteers to sift through a pre-filtered list of geo-located tweets to verify real-time aurora sightings. In this study, the current implementation of this crowdsourced verification system, including the process of geo-locating tweets, is described and its accuracy (which, overall, is found to be 68.4%) is determined. The findings suggest that citizen science volunteers are able to accurately filter out unrelated, spam-like, Twitter data but struggle when filtering out somewhat related, yet undesired, data. The citizen scientists particularly struggle with determining the real-time nature of the sightings and care must therefore be taken when relying on crowdsourced identification.

Item Type:
Journal Article
Journal or Publication Title:
Citizen Science: Theory and Practice
Subjects:
?? twittercrowdsourcingaurorasightingscitizen science ??
ID Code:
79750
Deposited By:
Deposited On:
25 May 2016 15:42
Refereed?:
Yes
Published?:
Published
Last Modified:
01 Oct 2024 00:15