Classecol: classifiers to understand public opinions of nature

Johnson, T. F., Kent, H., Hill, B. M., Dunn, G., Dommett, L., Penwill, N., Francis, T. and Gonzalez-Suarez, M. ORCID: https://orcid.org/0000-0001-5069-8900 (2021) Classecol: classifiers to understand public opinions of nature. Methods in Ecology and Evolution, 12 (7). pp. 1329-1334. ISSN 2041-210X doi: 10.1111/2041-210X.13596

Abstract/Summary

1) Human perceptions of nature, once the domain of the social sciences, are now an important part of environmental research. However, the data and tools to tackle this research are lacking or are difficult to apply. 2) Here, we present a collection of text classifier models to identify text relevant to the broad topics of hunting and nature, describing whether opinions are pro- or against-hunting, or show interest, concern, or dislike of nature. The methods also include a biographical classification – describing whether the author of the text is a person, nature expert, nature organisation, or ‘Other’. The classifiers were developed using an extensive social media dataset, and are designed to support qualitative analysis of big data (especially from Twitter). 3) The classifiers accurately identified biographies, text related to hunting and nature, and the stance towards hunting and nature (weighted F-scores: 0.79 - 0.99; 1 indicates perfect accuracy). 4) These classifiers, alongside an array of other text processing and analysis functions, are presented in the form of an R package classecol. classecol also acts as a proof of concept that nature related text classifiers can be developed with high accuracy.

Item Type	Article
URI	https://reading-clone.eprints-hosting.org/id/eprint/96456
Item Type	Article
Refereed	Yes
Divisions	Life Sciences > School of Biological Sciences > Ecology and Evolutionary Biology
Publisher	Wiley-Blackwell
Download/View statistics	View download statistics for this item