TechPrufer
Sign inGet started

Blogs›Apple›UMAP kNN Graph Analysis

UMAP kNN Graph Analysis

UMAP kNN Graph Analysis

1
posts
2026

This post introduces novel applications of graph algorithms (PageRank, k-core decomposition, clustering coefficient) to the internal kNN graph generated by UMAP. These methods enhance data sensemaking by identifying representative data points, revealing data density, and detecting similar data neighborhoods, offering practical and competitive alternatives to existing methods.

2026

Dimensionality Reduction Meets Network Science: Sensemaking on UMAP’s kNN Graph

7/30/2026

Introduces the application of PageRank for identifying representative data points, k-core decomposition for revealing dense core regions versus sparse periphery, and clustering coefficient for detecting tight-knit neighborhoods within UMAP's internal kNN graph. Evaluates these methods on MNIST and Fashion MNIST datasets.