How to interpret the iteration and the log likelihood statistics of the Parallel LDA output in Topic Modeling
Hello @khatiwadar18,
a similar question has already been answered here: number of iterations in the Topic Extractor node
In short, the log likelihood value for each iteration can be interpreted as a convergence score of the algorithm. Once the log likelihood does not change anymore, that means that these number of iterations are enough to extract the topics from the dataset.
Best regards
Jörg
1 Like
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.