We provide access to development code and data generated within our project.
The project source code is available on GitHub:
a Java library that can be used for LDA topic modeling (though our reported
work uses Mallet for this), hierarchical clustering, concept graph
generation (i.e., the prediction of concept dependency edges between
LDA topics), and the generation of reading lists based on a concept graph
and other inputs.
This is a
Python library and associated scripts that can run expand a base corpus with
relevant, pedaogically diverse documents, generate a JSON concept graph
(using techknacq-core for edge computation), and generate reading lists
given a JSON concept graph and a query.