01_Topic Detection Clustering

After classic pre-processing (in streaming mode!) we: - extract the document keywords with a KeyGraph Keyword Extractor node. - we display the words making the protottype of each cluster into a word cloud. The word clouds are displayed in the interactive view of component Cluster 0-9. Note. We limit the score of the keywords to 5, in order to avoid an uncontrolled growth of the number of columns feeding the clustering algorithm.

This is a companion discussion topic for the original entry at https://kni.me/w/roVBhnZejXO06B_W