Model Training & Adaptation
Clustering
Putting similar examples into groups based on their characteristics.
Example
A system groups customers with similar purchase patterns.
Why people use it
It helps reveal groups that were not labeled beforehand.
What you'll hear
“Which customers naturally seem to group together?”
What this means for you
Interpret and validate groups before assigning them business meaning.
Can you control it?
Developer-only
The people building or running the AI choose this setup. An everyday user generally needs their help to change how this part works.
Common questions
- Are clusters the same as predefined business segments?
- Not necessarily. Clusters are discovered from the selected description and method.
- Can changing the similarity rule change the groups?
- Yes. Grouping by spending can produce different results from grouping by location or visit frequency.
- Can a cluster contain very different people?
- Yes. Sharing measured characteristics does not make every person in a group alike.