Topic Extraction (Parallel LDA) Node question


just for confirmation... this node works only with english language, does it?

My best


LDA is not specific to any language, in that it merely finds terms that occur together in a pattern that is suggestive of a topic. It depends on the co-occurrence of words, not the meaning of them, so it should work reasonably well in any language. The challenge is likely to lie in how you pre-process the text, for example which stopwords you remove, how you deal with different tenses, capitalisation, etc.

Thank you, Sugna.