The United Nations’ Sustainable Development Goals (SDGs) have become an important guideline for higher-education and research institutions to monitor and plan their contributions to social, economic, and environmental transformations.
text2sdg package is the first open-source, multi-system analysis package that identifies SDGs in text, opening up the opportunity to monitor any type of text-based data, including scientific output and corporate publications.
text2sdg package is developed by Dirk U. Wulff and Dominik S. Meier, with contributions from Rui Mata and the Center for Cognitive and Decision Sciences. It is published under the GNU General Public License.
The current stable version is available on CRAN and can be installed via
The latest development version on GitHub can be installed via
devtools::install_github("dwulff/text2sdg"). Note that this requires prior installation of the
To identify SDGs in a series of documents, the user can choose between two approaches, an individual systems approach implementing six individual query systems and an ensemble approach powered by machine learning that integrates these systems. It is recommended to use the more accurate and bias-free ensemble approach (see Wulff, Meier, & Mata, 2023).
# vector of texts texts = c("This is text 1", "This is text 2") # individual systems approach hits = detect_sdg_systems(texts) # ensemble approach hits = detect_sdg(texts)
If you use the
text2sdg package for published work, we kindly ask that you cite the package as follows:
Meier, D. S., Mata, R., & Wulff, D. U. (2021). text2sdg: An open-source solution to monitoring sustainable development goals from text. arXiv. https://arxiv.org/abs/2110.05856
Depending on the use of the package, also consider referencing the related article below.
Wulff, Dirk U., Meier, Dominik S., & Mata, R. (2023). Using novel data and ensemble models to improve automated labeling of Sustainable Development Goals. arXiv. https://arxiv.org/abs/2301.11353