Shiny App to explore a scatterplot of tokens connected to a scatterplot of their context words. This includes information on HDBSCAN clusters of the tokens and the representativity of the context words in terms of these clusters. It's a 'fancier' version of Level 3 of this visualization. See here for a brief illustration.
Fork this repository, change the datasets in the "cloudspotting-medoids" subdirectory to your own data, adapt the variables in "R/variables.R" and run the Shiny App.
This work has been possible by the C1 KU Leuven funding of the Nephological Semantics project.