The Principal Components Analysis of a Graph, and its Relationships to Spectral Clustering

by: Marco S Francois

View FullText article

CiteSeer (at Penn State), CiteSeer (at MIT), CiteSeer (at Zurich), CiteSeer (at Singapore)

Reviews [Write a review of this article]

There are no reviews of this article

Find related articles from these CiteULike users

kevynct, xtizon, AbnerCYH, yinghua77, sugarexpletive, jiria, ehohman, pprett, CSBBGraphTheory (group), rchauhan, mamader

Find related articles with these CiteULike tags

algebra, analysis, clustering, complex, graph, graphs, graph-theory, hits, image-analysis, kpca, math, misc, pca, random, randomwalk, spectral, spectral-clustering, walk

Abstract

This work presents a novel procedure for computing (1) distances between nodes of a weighted, undirected, graph, called the Euclidean Commute Time Distance (ECTD), and (2) a subspace projection of the nodes of the graph that preserves as much variance as possible, in terms of the ECTD -- a principal components analysis of the graph. It is based on a Markov-chain model of random walk through the graph. The model assigns transition probabilities to the links between nodes, so that a random...

@misc{citeulike:165628, abstract = {This work presents a novel procedure for computing (1) distances between nodes of a weighted, undirected, graph, called the Euclidean Commute Time Distance (ECTD), and (2) a subspace projection of the nodes of the graph that preserves as much variance as possible, in terms of the ECTD -- a principal components analysis of the graph. It is based on a Markov-chain model of random walk through the graph. The model assigns transition probabilities to the links between nodes, so that a random...}, author = {Francois, Marco S. }, citeulike-article-id = {165628}, keywords = {graphs}, posted-at = {2008-05-02 07:49:17}, priority = {2}, title = {The Principal Components Analysis of a Graph, and its Relationships to Spectral Clustering}, url = {http://citeseer.ist.psu.edu/704731.html} }

RIS record

TY - GEN ID - citeulike:165628 L3 - citeulike-article-id:165628 TI - The Principal Components Analysis of a Graph, and its Relationships to Spectral Clustering N2 - This work presents a novel procedure for computing (1) distances between nodes of a weighted, undirected, graph, called the Euclidean Commute Time Distance (ECTD), and (2) a subspace projection of the nodes of the graph that preserves as much variance as possible, in terms of the ECTD -- a principal components analysis of the graph. It is based on a Markov-chain model of random walk through the graph. The model assigns transition probabilities to the links between nodes, so that a random... KW - graphs AU - Francois, Marco UR - http://citeseer.ist.psu.edu/704731.html ER -

RIS BibTeX