skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: DELTACON: A Principled Massive-Graph Similarity Function with Attribution

Abstract

How much did a network change since yesterday? How different is the wiring between Bob's brain (a left-handed male) and Alice's brain (a right-handed female)? Graph similarity with known node correspondence, i.e. the detection of changes in the connectivity of graphs, arises in numerous settings. In this work, we formally state the axioms and desired properties of the graph similarity functions, and evaluate when state-of-the-art methods fail to detect crucial connectivity changes in graphs. We propose DeltaCon, a principled, intuitive, and scalable algorithm that assesses the similarity between two graphs on the same nodes (e.g. employees of a company, customers of a mobile carrier). In our experiments on various synthetic and real graphs we showcase the advantages of our method over existing similarity measures. We also employ DeltaCon to real applications: (a) we classify people to groups of high and low creativity based on their brain connectivity graphs, and (b) do temporal anomaly detection in the who-emails-whom Enron graph.

Authors:
 [1];  [1];  [2];  [3];  [1]
  1. Carnegie Mellon Univ., Pittsburgh, PA (United States). Computer Science Dept.
  2. Duke Univ., Durham, NC (United States). Dept. of Statistical Science
  3. Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Publication Date:
Research Org.:
Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1343040
Report Number(s):
LLNL-JRNL-677691
Journal ID: ISSN 1556-4681
Grant/Contract Number:  
AC52-07NA27344
Resource Type:
Journal Article: Accepted Manuscript
Journal Name:
ACM Transactions on Knowledge Discovery from Data
Additional Journal Information:
Journal Volume: 10; Journal Issue: 3; Conference: SIAM International Conference on Data Mining, Austin, Texas, USA, 5/02/2013 - 5/04/2013; Journal ID: ISSN 1556-4681
Publisher:
Association for Computing Machinery
Country of Publication:
United States
Language:
English
Subject:
99 GENERAL AND MISCELLANEOUS; 97 MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE; algorithms; experimentation; graph similarity; graph comparison; anomaly detection; network monitoring; graph classification; node attribution; edge attribution

Citation Formats

Koutra, Danai, Shah, Neil, Vogelstein, Joshua T., Gallagher, Brian, and Faloutsos, Christos. DELTACON: A Principled Massive-Graph Similarity Function with Attribution. United States: N. p., 2014. Web. doi:10.1145/2824443.
Koutra, Danai, Shah, Neil, Vogelstein, Joshua T., Gallagher, Brian, & Faloutsos, Christos. DELTACON: A Principled Massive-Graph Similarity Function with Attribution. United States. https://doi.org/10.1145/2824443
Koutra, Danai, Shah, Neil, Vogelstein, Joshua T., Gallagher, Brian, and Faloutsos, Christos. Thu . "DELTACON: A Principled Massive-Graph Similarity Function with Attribution". United States. https://doi.org/10.1145/2824443. https://www.osti.gov/servlets/purl/1343040.
@article{osti_1343040,
title = {DELTACON: A Principled Massive-Graph Similarity Function with Attribution},
author = {Koutra, Danai and Shah, Neil and Vogelstein, Joshua T. and Gallagher, Brian and Faloutsos, Christos},
abstractNote = {How much did a network change since yesterday? How different is the wiring between Bob's brain (a left-handed male) and Alice's brain (a right-handed female)? Graph similarity with known node correspondence, i.e. the detection of changes in the connectivity of graphs, arises in numerous settings. In this work, we formally state the axioms and desired properties of the graph similarity functions, and evaluate when state-of-the-art methods fail to detect crucial connectivity changes in graphs. We propose DeltaCon, a principled, intuitive, and scalable algorithm that assesses the similarity between two graphs on the same nodes (e.g. employees of a company, customers of a mobile carrier). In our experiments on various synthetic and real graphs we showcase the advantages of our method over existing similarity measures. We also employ DeltaCon to real applications: (a) we classify people to groups of high and low creativity based on their brain connectivity graphs, and (b) do temporal anomaly detection in the who-emails-whom Enron graph.},
doi = {10.1145/2824443},
url = {https://www.osti.gov/biblio/1343040}, journal = {ACM Transactions on Knowledge Discovery from Data},
issn = {1556-4681},
number = 3,
volume = 10,
place = {United States},
year = {2014},
month = {5}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Citation Metrics:
Cited by: 30 works
Citation information provided by
Web of Science

Save / Share:

Works referenced in this record:

Graph based anomaly detection and description: a survey
journal, July 2014


Authoritative sources in a hyperlinked environment
journal, September 1999


Graph Comparison Using Fine Structure Analysis
conference, August 2010


Fast anomaly detection despite the duplicates
conference, January 2013


Thirty Years of Graph Matching in Pattern Recognition
journal, May 2004


Graph-based anomaly detection
conference, January 2003


Anomaly detection in dynamic networks: a survey
journal, March 2015


BIG-ALIGN: Fast Bipartite Graph Alignment
conference, December 2013

  • Koutra, Danai; Tong, Hanghang; Lubensky, David
  • 2013 IEEE International Conference on Data Mining (ICDM), 2013 IEEE 13th International Conference on Data Mining
  • https://doi.org/10.1109/ICDM.2013.152

TimeCrunch: Interpretable Dynamic Graph Summarization
conference, January 2015


Topic-sensitive pagerank: A context-sensitive ranking algorithm for web search
journal, July 2003


Web graph similarity for anomaly detection
journal, February 2010


Eigenspace-based anomaly detection in computer systems
conference, January 2004


Network similarity via multiple social theories
conference, January 2013

  • Berlingerio, Michele; Koutra, Danai; Eliassi-Rad, Tina
  • Proceedings of the 2013 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining - ASONAM '13
  • https://doi.org/10.1145/2492517.2492582

Revealing the Hidden Language of Complex Networks
journal, April 2014


Shortest-Path Kernels on Graphs
conference, January 2005


A Set of Measures of Centrality Based on Betweenness
journal, March 1977


Algebraic connectivity of graphs [Algebraic connectivity of graphs]
journal, January 1973


Propagation of trust and distrust
conference, January 2004


SNARE: a link analytic system for graph labeling and risk detection
conference, January 2009

  • McGlohon, Mary; Bay, Stephen; Anderle, Markus G.
  • Proceedings of the 15th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '09
  • https://doi.org/10.1145/1557019.1557155

The anatomy of a large-scale hypertextual Web search engine
journal, April 1998


Cyclic pattern kernels for predictive graph mining
conference, January 2004

  • Horváth, Tamás; G?rtner, Thomas; Wrobel, Stefan
  • Proceedings of the 2004 ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '04
  • https://doi.org/10.1145/1014052.1014072

Locality Statistics for Anomaly Detection in Time Series of Graphs
journal, February 2014


Graph kernels based on tree patterns for molecules
journal, October 2008


A study of graph spectra for comparing graphs and trees
journal, September 2008


Personalized PageRank vectors for tag recommendations: inside FolkRank
conference, January 2011


MultiAspectForensics: Pattern Mining on Large-Scale Heterogeneous Networks with Tensor Analysis
conference, July 2011


Temporal Scale of Processes in Dynamic Networks
conference, December 2011

  • Caceres, Rajmonda Sulo; Berger-Wolf, Tanya; Grossman, Robert
  • 2011 IEEE International Conference on Data Mining Workshops (ICDMW), 2011 IEEE 11th International Conference on Data Mining Workshops
  • https://doi.org/10.1109/ICDMW.2011.165

VOG: Summarizing and Understanding Large Graphs
conference, April 2014


On the evolution of user interaction in Facebook
conference, January 2009


Fast Random Walk Graph Kernel
conference, December 2013


Graph evolution: Densification and shrinking diameters
journal, March 2007


Locality Sensitive Outlier Detection: A ranking driven approach
conference, April 2011

  • Wang, Ye; Parthasarathy, Srinivasan; Tatikonda, Shirish
  • 2011 IEEE International Conference on Data Engineering (ICDE 2011), 2011 IEEE 27th International Conference on Data Engineering
  • https://doi.org/10.1109/ICDE.2011.5767852

Visual Graph Comparison
conference, July 2009


D elta C on : A Principled Massive-Graph Similarity Function
conference, December 2013


A measure of betweenness centrality based on random walks
journal, January 2005


SimRank: a measure of structural-context similarity
conference, January 2002


Comparison of graphs by their number of spanning trees
journal, November 1976


Fast computation of SimRank for static and dynamic information networks
conference, January 2010


Gelling, and melting, large graphs by edge manipulation
conference, January 2012

  • Tong, Hanghang; Prakash, B. Aditya; Eliassi-Rad, Tina
  • Proceedings of the 21st ACM international conference on Information and knowledge management - CIKM '12
  • https://doi.org/10.1145/2396761.2396795

Metric forensics: a multi-level approach for mining volatile graphs
conference, January 2010

  • Henderson, Keith; Eliassi-Rad, Tina; Faloutsos, Christos
  • Proceedings of the 16th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '10
  • https://doi.org/10.1145/1835804.1835828

More is simpler: effectively and efficiently assessing node-pair similarities based on hyperlinks
journal, September 2013


Magnetic Resonance Connectome Automated Pipeline: An Overview
journal, March 2012


TensorSplat: Spotting Latent Anomalies in Time
conference, October 2012


Interactive graph matching and visual comparison of graphs and clustered graphs
conference, January 2012


Weighted graph comparison techniques for brain connectivity analysis
conference, January 2013


OPAvion: mining and visualization in large graphs
conference, January 2012


Localizing anomalous changes in time-evolving graphs
conference, January 2014


Visual comparison for information visualization
journal, September 2011


CopyCatch: stopping group attacks by spotting lockstep behavior in social networks
conference, January 2013


Aligning graphs and finding substructures by a cavity approach
journal, February 2010


Linearized and single-pass belief propagation
journal, January 2015


MIGRAINE: MRI Graph Reliability Analysis and Inference for Connectomics
conference, December 2013


Message-Passing Algorithms for Sparse Network Alignment
journal, March 2013


Graph-based anomaly detection
conference, January 2003


    Works referencing / citing this record:

    Fast network discovery on sequence data via time-aware hashing
    journal, December 2018


    Automated assessment of knowledge hierarchy evolution: comparing directed acyclic graphs
    journal, December 2018


    Non-backtracking cycles: length spectrum theory and graph mining applications
    journal, June 2019


    Unsupervised network embeddings with node identity awareness
    journal, October 2019


    Wisdom of stakeholder crowds in complex social–ecological systems
    journal, January 2020


    Hierarchical Change Point Detection on Dynamic Networks
    conference, June 2017


    Graph Summarization Methods and Applications: A Survey
    journal, July 2018


    Discovering and deciphering relationships across disparate data modalities
    journal, January 2019