The question behind the work.
What structure can be recovered without using labels, and how sensitive is that structure to a distance metric?
Dataset & environment
Academic clustering explorations; no assignment dataset has been attached to this portfolio.
From the idea to the method.
- Scale continuous features and inspect pairwise distances.
- Compare linkage criteria and examine a dendrogram.
- Evaluate several cluster counts using silhouette scores and qualitative inspection.
What this study explores.
- Academic focus: hierarchical clustering and distance-based analysis.
- Interpreting dendrograms and comparing cluster separation.
How to assess it carefully.
Use silhouette analysis, stability under resampling, and domain interpretation rather than treating a single score as proof.
What is documented.
Experimental artifacts and verified numerical results have not been published. The interactive visuals explain the method; they are not benchmark results.
Training curves, accuracy comparisons, and repository links will be added alongside the corresponding artifacts. No unverified benchmark numbers are shown.
A foundation for the next experiment.
- Clusters depend on the feature representation and the definition of distance.
Future updates can include code, screenshots, reproducibility instructions, and verified outcomes.