Robust outlier detection using commute time and eigenspace embedding

Nguyen Lu Dang Khoa, Sanjay Chawla

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

16 Scopus citations


We present a method to find outliers using 'commute distance' computed from a random walk on graph. Unlike Euclidean distance, commute distance between two nodes captures both the distance between them and their local neighborhood densities. Indeed commute distance is the Euclidean distance in the space spanned by eigenvectors of the graph Laplacian matrix. We show by analysis and experiments that using this measure, we can capture both global and local outliers effectively with just a distance based method. Moreover, the method can detect outlying clusters which other traditional methods often fail to capture and also shows a high resistance to noise than local outlier detection method. Moreover, to avoid the O(n3) direct computation of commute distance, a graph component sampling and an eigenspace approximation combined with pruning technique reduce the time to O(nlogn) while preserving the outlier ranking.

Original languageEnglish
Title of host publicationAdvances in Knowledge Discovery and Data Mining - 14th Pacific-Asia Conference, PAKDD 2010, Proceedings
Number of pages13
EditionPART 2
StatePublished - 2010
Externally publishedYes
Event14th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2010 - Hyderabad, India
Duration: Jun 21 2010Jun 24 2010

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 2
Volume6119 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference14th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2010


  • Commute distance
  • Eigenspace embedding
  • Nearest neighbor graph
  • Outlier detection
  • Random walk


Dive into the research topics of 'Robust outlier detection using commute time and eigenspace embedding'. Together they form a unique fingerprint.

Cite this