# Clustering on 3,500 Beatport EDM tracks (full)

Source: [Clustering on 3,500 Beatport EDM tracks (full)](<https://www.sota2.com/research/sota/clustering-on-3-500-beatport-edm-tracks-full>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-31T06:59:55.151Z

Evaluation of clustering performance on the full set of 3,500 Beatport EDM tracks, utilizing internal consistency and external validation metrics like Silhouette, ARI, and NMI.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Clustering |
| Dataset (including any stated split) | 3,500 Beatport EDM tracks (full) |
| Metric | Silhouette Score |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Silhouette Score](<https://www.sota2.com/research/sota/clustering-on-3-500-beatport-edm-tracks-full.md?metric=Silhouette%20Score>): Higher is better; unit: SCALAR.

- [Davies-Bouldin Index](<https://www.sota2.com/research/sota/clustering-on-3-500-beatport-edm-tracks-full.md?metric=Davies-Bouldin%20Index>): Lower is better; unit: SCALAR.

- [Cophenetic Correlation](<https://www.sota2.com/research/sota/clustering-on-3-500-beatport-edm-tracks-full.md?metric=Cophenetic%20Correlation>): Higher is better; unit: SCALAR.

- [Adjusted Rand Index (ARI)](<https://www.sota2.com/research/sota/clustering-on-3-500-beatport-edm-tracks-full.md?metric=Adjusted%20Rand%20Index%20(ARI)>): Higher is better; unit: SCALAR.

- [Normalized Mutual Information (NMI)](<https://www.sota2.com/research/sota/clustering-on-3-500-beatport-edm-tracks-full.md?metric=Normalized%20Mutual%20Information%20(NMI)>): Higher is better; unit: SCALAR.

- [Mutual Information Score](<https://www.sota2.com/research/sota/clustering-on-3-500-beatport-edm-tracks-full.md?metric=Mutual%20Information%20Score>): Higher is better; unit: SCALAR.

- [Purity Score](<https://www.sota2.com/research/sota/clustering-on-3-500-beatport-edm-tracks-full.md?metric=Purity%20Score>): Higher is better; unit: SCALAR.

- [Cluster Balance](<https://www.sota2.com/research/sota/clustering-on-3-500-beatport-edm-tracks-full.md?metric=Cluster%20Balance>): Higher is better; unit: SCALAR.

- [Normalized Balance](<https://www.sota2.com/research/sota/clustering-on-3-500-beatport-edm-tracks-full.md?metric=Normalized%20Balance>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Silhouette Score | Result date | Report paper |
| --- | --- | --- | --- |
| K-Means | 0.0956 | Not recorded | 68fa7b561e209131cea87e00 |
| Divisive Hierarchical Clustering | 0.0666 | Not recorded | 68fa7b561e209131cea87e00 |
| CLAP-tiny | 0.0496 | Not recorded | 68fa7b561e209131cea87e00 |
| MERT-95M | 0.0491 | Not recorded | 68fa7b561e209131cea87e00 |

### K-Means

Silhouette Score: 0.0956. Result ID: 69cb706947f34ce924b337bc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b561e209131cea87e00.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b561e209131cea87e00.md>)

Recorded evaluation configuration: {"Feature Source":"Extracted Features","k (cluster count)":"35","Initialization":"K-means++"}. This record does not assert independent reproduction.

### Divisive Hierarchical Clustering

Silhouette Score: 0.0666. Result ID: 69cb706947f34ce924b337bd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b561e209131cea87e00.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b561e209131cea87e00.md>)

Recorded evaluation configuration: {"Feature Source":"Extracted Features","k (cluster count)":"35"}. This record does not assert independent reproduction.

### CLAP-tiny

Silhouette Score: 0.0496. Result ID: 69cb706947f34ce924b337bf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b561e209131cea87e00.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa729e1e209131cea57959.md>)

Recorded evaluation configuration: {"Feature Source":"Pre-trained Embeddings","Clustering Algorithm":"K-Means","k (cluster count)":"35"}. This record does not assert independent reproduction.

### MERT-95M

Silhouette Score: 0.0491. Result ID: 69cb706947f34ce924b337be.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b561e209131cea87e00.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa766e1e209131cea6cf38.md>)

Recorded evaluation configuration: {"Feature Source":"Pre-trained Embeddings","Clustering Algorithm":"K-Means","k (cluster count)":"35"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
