Similar to what's already present in the hype-detector branch, but more bottom-up: instead of specifiying a search term, find the highest scoring terms per time span.
For instance, concatenate texts per time span and treat them as documents (labeled by their date).
mariekevh
commented
6 years ago
Similar to what's already present in the hype-detector branch, but more bottom-up: instead of specifiying a search term, find the highest scoring terms per time span. For instance, concatenate texts per time span and treat them as documents (labeled by their date).
see also https://stackoverflow.com/questions/34232190/scikit-learn-tfidfvectorizer-how-to-get-top-n-terms-with-highest-tf-idf-score