Examples
Examples
Try the starting example
- JSON Points [ID,equal-length numeric feature vector,cluster label]
- [["A",[0,0],"X"],["B",[0,2],"X"],["C",[10,0],"Y"],["D",[10,2],"Y"]]
- Per-point Euclidean silhouette
- [["A",0.801960972814],["B",0.801960972814],["C",0.801960972814],["D",0.801960972814]]
- Mean silhouette
- 0.801961
Try another scenario
- JSON Points [ID,equal-length numeric feature vector,cluster label]
- [["A",[0,0],"X"],["B",[0,4],"X"],["C",[10,0],"Y"],["D",[10,4],"Y"]]
- Per-point Euclidean silhouette
- [["A",0.614835192865],["B",0.614835192865],["C",0.614835192865],["D",0.614835192865]]
- Mean silhouette
- 0.614835
Calculation method
Calculation method
Per-point Euclidean silhouette
a=mean distance to other own-cluster points; b=min other-cluster mean distance; s=(b−a)/max(a,b), zero for singleton or a=b=0.
Mean silhouette
Arithmetic mean across point silhouettes.
Symbols and units
- a
- JSON Points [ID,equal-length numeric feature vector,cluster label]
Use the units shown. A list keeps the order in which you enter its values. The symbol ^ means “raised to the power of”.
Assumptions & limits
Assumptions & limits
- Euclidean distance on already scaled features; exact pairwise computation. Metric choice changes interpretation; no cluster truth or statistical significance claimed. Supported limits: 3–200 points, 1–20 finite features. Singleton silhouette defined as zero.
Are these inputs private?
These calculations run in your browser. A shared link contains the values you choose to share. Only questions submitted to the separate AI finder are sent to OpenAI.
Further reading
Background reading; this does not imply endorsement or independent review of SumFizz.