Predictor belongs to the cluster

elena2020chaoelena2020chao Member Posts: 13
edited December 2018 in Help

Hello
How after clustering kmeans on texts
Identify, in which cluster is the new entered text?
How do I predict which cluster is the new text?
Thankful
And how after text clustering with kmeans, a picture like the following is plotted?
What is the word cluster center?

۲.JPG

Answers

  • kypexinkypexin Moderator, RapidMiner Certified Analyst, Member Posts: 194   Unicorn

    Hi @elena2020chao

     

    For your first question: 

     

    Not sure why you explicitly need to predict a cluster.  

     

    Clustering algorithm adds cluster labels to the data, then it is possible to built for example a decision tree model over these labels, this way you can predict a cluster label for the new text. 

     

    But also consider the simplest and most straightforward way (especially if you don't have many texts), in which you just add a new text to your corpus and then run clustering algorithm over it once again, this way new text will be assigned to one of the clusters.

Sign In or Register to comment.