A new paper published on arXiv proposes the Centroid Index (CI) as a recommended method for evaluating clustering when ground truth data is available. The paper reviews common external validity indexes, particularly those based on set-matching measures. For more granular, point-level evaluation, the Pair-set index (PSI) is suggested for its normalized score that is not influenced by cluster sizes. If equal weighting of all points is desired, clustering accuracy (ACC) or similar set-matching measures are deemed suitable. AI
IMPACT Provides new evaluation metrics for clustering algorithms, potentially improving the development and assessment of AI models that rely on clustering.
RANK_REASON The cluster contains a research paper detailing new methods for evaluating clustering algorithms.
- alphaXiv
- arXiv
- CatalyzeX
- Centroid Index (CI)
- Clustering accuracy (ACC)
- Connected Papers
- CORE Recommender
- DagsHub
- Gotit.pub
- Hugging Face
- Influence Flower
- Litmaps
- Pair-set index (PSI)
- ScienceCast
- scite Smart Citations
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
