/knowledge

Primary LanguageJupyter NotebookGNU General Public License v3.0GPL-3.0

knowledge

Plan to use https://babelnet.org to identify "entities."

First pass will be naive for loop searching for bi-grams / n-grams.

End goal is to classify each "state-of-the-state" speech by some large overarching "topic(s)" or "theme(s)"

In between First Pass and the End Goal is a lot of code magic.