This workflow explores a possible way to identify the optimal number of topics which describe appropiately a corpus.
After choosing the best solution, it also creates a visual representation of them in a lower dimensional space by computing a distance between topics.
Workflow
Identify and visually represent a topic model
Used extensions & nodes
Created with KNIME Analytics Platform version 4.6.3 Note: Not all extensions may be displayed.
- Go to item
KNIME Textprocessing - Deeplearning4J Integration (64bit only)
KNIME AG, Zurich, Switzerland
Version 4.5.0
- Go to item
- Go to item
- Go to item
- Go to item
- Go to item
- Go to item
Legal
By using or downloading the workflow, you agree to our terms and conditions.