What term describes the phenomenon where distance-based methods like K-means lose effectiveness as the number of features increases?

What term describes the phenomenon where distance-based methods like K-means lose effectiveness as the number of features increases?

• A. Curse of Dimensionality
• B. Overfitting
• C. High-Dimensional Scaling
• D. Dimensional Instability

Answer: (A)

Distance-based methods rely on meaningful differences in distance between points. As the number of features grows, data become sparse in the feature space and distances lose contrast, so points end up almost equally far from one another. This is the curse of dimensionality. For K-means, which groups points by proximity to centroids using those distances, the lack of discriminative distance makes it hard to identify distinct clusters—the distances to different centroids become similar, and clustering becomes unreliable. You often need dimensionality reduction or alternative distance metrics to restore effectiveness. Overfitting describes a model fitting noise with too much complexity, not this deterioration of distance usefulness in high dimensions, and the other terms aren’t standard descriptors of this phenomenon.

Distance-based methods rely on meaningful differences in distance between points. As the number of features grows, data become sparse in the feature space and distances lose contrast, so points end up almost equally far from one another. This is the curse of dimensionality. For K-means, which groups points by proximity to centroids using those distances, the lack of discriminative distance makes it hard to identify distinct clusters—the distances to different centroids become similar, and clustering becomes unreliable. You often need dimensionality reduction or alternative distance metrics to restore effectiveness. Overfitting describes a model fitting noise with too much complexity, not this deterioration of distance usefulness in high dimensions, and the other terms aren’t standard descriptors of this phenomenon.