Model Training & Adaptation
Hierarchical clustering
Organizing similar items into groups within larger groups, like branches of a family tree.
Example
Documents form small topic groups that are combined into broader groups.
Why people use it
It helps people see both broad groups and smaller groups within them.
What you'll hear
“These small clusters belong to a larger group.”
What this means for you
Inspect representative items before interpreting the hierarchy.
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
- Does the hierarchy have to follow a real-world family tree?
- No. The branches reflect the selected similarity method, not necessarily a natural or historical relationship.
- Must I choose one final number of groups immediately?
- Not always. The hierarchy can be examined at different levels of detail.
- Can the choice of similarity change the hierarchy?
- Yes. Different ways of measuring likeness can produce different groupings.