0

I tried creating a topic modelling using pyldavis gensim library and now the clusters are made. I have one question about the same.Clusters made are cut from the edges. Is there a problem or its fine?

A similar sample graph would be: https://miro.medium.com/max/1086/1*80r1wLMCyK615S6-j3TLWQ.png. Also, is there a citation that can back the result?

Thanks I tried pyldavis for the same.

Shashwat Shankar
  • 3
  • 3
  • Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. – Community Apr 01 '22 at 18:02
  • The `pyLDAvis` & `gensim` libraries are separate things, & it's unclear what you mean by "Clusters made are cut from the edges" without more details. What are you seeing, and what did you expect to see instead? – gojomo Apr 01 '22 at 18:34
  • I have used pyldavis library for making "Intertopic Distance Map Map (via multidimensional scaling)" {LdaModel()} topic modelling visualisation graph (using Gensim package) with 4 clusters. The 4 clusters, even though completely non-overlapping, one of the cluster is out of the graph in such a way that the circle is cut to show only a part of cluster. So I wonder if that's ok or do we need to have all clusters completely disjointed and visible. A similar sample graph would be: https://miro.medium.com/max/1086/1*80r1wLMCyK615S6-j3TLWQ.png. Also, is there a citation that can back the result? – Shashwat Shankar Apr 02 '22 at 17:03
  • I'm not sure what on your linked image fits the description "cut to only show part of a cluster". – gojomo Apr 05 '22 at 01:45
  • The cluster named '2' has a portion of it's left side of the circle hidden/cut at the end of the graph. – Shashwat Shankar Apr 06 '22 at 14:22

0 Answers0