Re:Document Clustering with Python
by strong_ai on Friday October 16, 2015 @07:36PM
(#10402670)
It would be really interesting to see a visualization (maybe using PCA) of the LDA vectors for each document. The topics are not super convincing that the LDA approach worked well.
Other than that, this is a good intro to NLP and calculating document similarity. Well done!
Re:Document Clustering with Python
by stuartaxelowen on Saturday October 17, 2015 @05:15PM
(#10406142)
I'm very interested in ways we can expand LDA (and other topic models) to retain more of the meaning of the documents, especially for small feedback (like reviews), such that a human could explore the results and find impactful, actionable data.
Using N-grams increases the dimensionality way too much.
I've thought about word-vectors as an option, but am unsure how similar terms could be grouped internally in LDA.
This kind of subject is probably worth a PHD thesis, honestly.
Other than that, this is a good intro to NLP and calculating document similarity. Well done!
Using N-grams increases the dimensionality way too much.
I've thought about word-vectors as an option, but am unsure how similar terms could be grouped internally in LDA.
This kind of subject is probably worth a PHD thesis, honestly.