Published March 1, 2021 | Version v1
Dataset Open

Annotation of press articles with topic modelling and naïve Bayes classifiers

Creators

Description

Annotation of historic press articles, organised per titles, with topic modelling and naïve Bayes classifiers (NBC2+4).

List of the 100 most significant words for the 4th round of NBC for the antimodern conception of Europe.

List of topics with labels and distribution per title, trained on the articles collection of each selected Swiss press title.

Files

Files (59.6 MB)

Name Size Download all
md5:cd6d91bbd30a7a847f01a400004920d9
59.5 MB Download
md5:96b4b5d52eab1a39be6eb9cfcca58c9a
39.8 kB Download