For those new to text analysis, it can sound intriguing, but identifying and implementing a project can be challenging. Below are some ways you could use Constellate and text analysis.
Teaching Use Cases
- Northeastern University regularly runs a class entitled "Python and Text Analysis for Absolute Beginners" (see their August 2021 calendar event.) This two hour class draws registrants from across disciplines. Students use the Constellate dataset builder to create datasets of interest to them and then work on these datasets in the Constellate Lab using Jupyter Notebooks created by Northeastern staff.
- A number of schools run sessions introducing Constellate to their community that leverage the Constellate Jupyter Notebooks, including this workshop at Notre Dame and this workshop at the University of Pittsburgh.
- The Text Analysis Pedagogy Institute (tapi), an open educational institute funded by the NEH for the benefit of teachers (and aspiring teachers) of text analysis in the digital humanities uses the Constellate Lab (Jupyter) exclusively to run the classes developed by the instructors.
Research Use Cases
We regularly receive requests for research datasets, but to date none of this research from Constellate content is formally published. Thus, below we list some other text analysis research as examples.
There is considerable use of text-analysis in the sciences, including medicine. Below is an example where text analysis is being used to analyze internet forums to determine if it would be a useful technique for identifying trends in recreational drug use:
Matthijs Blankers, Daan van der Gouwe, Margriet van Laar,
4-Fluoramphetamine in the Netherlands: Text-mining and sentiment analysis of internet forums,
International Journal of Drug Policy, Volume 64, 2019, Pages 34-39,
Text-analysis can be used as the basis for the development of interactive, applications. For example, this interactive topic model of Signs was built to celebrate 40 years of publication in 2014:
Goldstone, Andrew, Susana Galán, C. Laura Lovin, Andrew Mazzaschi, and Lindsey Whitmore. 2014. An Interactive Topic Model of Signs, edited by Andrew Goldstone. Signs at 40. http://signsat40.signsjournal.org/topic-model.
When presented with easy-to-use visualizations, even those not expert in text-analysis can use results in their research. In the book chapter referenced below, the author uses a visualization (that could be approximated in Constellate) to show that the term 'entrepreneur' is not used much by nature conservationists.
Judith C. Jobse et al., “Chapter 10: Preparing a New Generation of Wilderness Entreperneurs,” in Rewilding European Landscapes, ed. Henrique M. Pereira and Laetitia Navarro (SpringerOpen, 2015), https://link.springer.com/book/10.1007%2F978-3-319-12039-3.