Segmentation in R

Hi!

I'm working on a report of a dataset that includes 1000 consumers. We have asked them about their knowledge on a certain topic, and I would love to create some persona's to refer to in the report. Its about sustainability, so for example the 'super sustainable' persona and the 'i dont care about sustainability' persona.

I read about several methods on clustering, and unsupervised machine learning caught my attention - specifically the elbow method. However, I did not find any documentation on how to do this in R. I'm not super experienced in R, and Chat GPT only confused me :wink:

Is there anyone who could help me with this, or do you have any how-to's available? Also please let me know if there is more information necessary on this topic.

Thanks in advance!

Just a point on terminology: "unsupervised machine learning" is typically a synonym of "clustering", and "the elbow method", that depends on the context but it's typically about choosing a number of cluster based on some metric, within a clustering method.

That is a big question, there are many many clustering methods (and many many R packages available for that). Some methods work better than others for each dataset. A good starting point may be k-means and hierarchical clustering, that are commonly used on many type of data. There is a lot available on the web, for example this book chapter gives theoretical explanations and code.

1 Like

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.