Close ×

Search EBBA


Visualization of ballad woodcut images, clustered into different groups by unsupervised machine vision methods. Each image was processed with an Inception Convolutional Neural Network, trained on ImageNet 2012, and projected into a two-dimensional manifold with the UMAP algorithm such that similar images appear proximate to one another.
A tool for visualizing the Latent Dirichlet Allocation Topic Model of the archive. Allows the user to view the word clusters that comprise the model as well the similarity of the topic clusters to each other.
Visualization of number of black-letter versus white-letter ballads in archive plotted by year. Interactive graph lets you view only black-letter ballads, only white-letter ballads, or both simultaneously.
Visualization of number of ballads in archive with tune imprints versus number of ballads in archive without tune imprints plotted by year and type of ballad. Interactive graph lets you view only ballads with tune imprints, only ballads without tune imprints, or both simultaneously.
Under development site that provides an interactive, mapped timeline of ballad seller shop locations in London. The current dataset is a sampling of about sixty booksellers working between 1550-1700.