Subclustering is the process of breaking existing group/clusters of data into smaller ones based on their similarities, trends and existing patterns. For instance a cluster on
water leakage can be broken down into sub-clusters such as
leakage in balcony,
leakage in bathroom,
leakage in the kitchen, etc.
Why subclustering can be beneficial?
Subclustering helps analyse data in a deeper and more fine-grained way. An example would be having a main cluster on all insurance claims and breaking it down into claims involving cars, houses, fire, flood, ect. Such a break-down can make data analysis much easier and more insightful.
Relevance AI's platform provides you with a no-code workflow to subcluster your clustered data with a few clicks.
Once you have uploaded and clustered your data, select your dataset and click on Sub-cluster under Workflows. An easy way to access a workflow is to search for it under Browse Workflows as shown in the image below.
The workflow wizard will open where you can
Select the field to be used for sub- clustering. This is done using AI which required vectors. If you have already clustered you data, vectors are generated and saved under your account. So, simply select the field whose name matches the free-text field from the drop down menu.
Specify to how many subclusters you wish to break the original clusters.
Select the parent clustering field. Clustering results are accessible under
AI categories (clusters)
Specify which Cluster to be subclustered
- Set up some filters to apply sub-clustering to a subset of your dataset
- If you wish to receive an email upon the workflow completion
- Specify if the change has to be applied to the newly uploaded entries or the whole dataset. This item only applies if you add new entries to an existing dataset
When done with the set up, click on "Run workflow". You will receive an email notification upon workflow completion (unless you specify otherwise under the optional settings).
Check out the instruction on how to view subclustering results on Explorer.
Updated 4 months ago