Help to get better results with the LDA Algorithm

Hi I am usinng the LDA Node to extract topics from my documents.
Before I can do this I have to preprocess my documents (punctuation, stemming etc.).
This is the result of my LDA Node.

As you can see the topics in the result are the stemmed words.

Is there a way to match the stemmed words with the corresponding noun of said stem?

For example that the LDA produces apple instead of apll.

Somebody can help?

You can use the Stanford Lemmatizer (with a POS Tagger node preceding) to help with this - but note that it only supports English.

1 Like