Skip to Main Content
HBS Home
  • About
  • Academic Programs
  • Alumni
  • Faculty & Research
  • Baker Library
  • Giving
  • Harvard Business Review
  • Initiatives
  • News
  • Recruit
  • Map / Directions
Faculty & Research
  • Faculty
  • Research
  • Featured Topics
  • Academic Units
  • …→
  • Harvard Business School→
  • Faculty & Research→
Publications
Publications
  • Article
  • Journal of Machine Learning Research

Fast Generalized Subset Scan for Anomalous Pattern Detection

By: Edward McFowland III, Skyler Speakman and Daniel B. Neill
  • Format:Electronic
ShareBar

Abstract

We propose Fast Generalized Subset Scan (FGSS), a new method for detecting anomalous patterns in general categorical data sets. We frame the pattern detection problem as a search over subsets of data records and attributes, maximizing a nonparametric scan statistic over all such subsets. We prove that the nonparametric scan statistics possess a novel property that allows for efficient optimization over the exponentially many subsets of the data without an exhaustive search, enabling FGSS to scale to massive and high-dimensional data sets. We evaluate the performance of FGSS in three real-world application domains (customs monitoring, disease surveillance, and network intrusion detection), and demonstrate that FGSS can successfully detect and characterize relevant patterns in each domain. As compared to three other recently proposed detection algorithms, FGSS substantially decreased run time and improved detection power for massive multivariate data sets.

Keywords

Pattern Detection; Anomaly Detection; Knowledge Discovery; Bayesian Networks; Scan Statistics

Citation

McFowland III, Edward, Skyler Speakman, and Daniel B. Neill. "Fast Generalized Subset Scan for Anomalous Pattern Detection." Art. 12. Journal of Machine Learning Research 14 (2013): 1533–1561.
  • Read Now

About The Author

Edward McFowland III

Technology and Operations Management
→More Publications

More from the Authors

    • 2023
    • Journal of the American Statistical Association

    Estimating Causal Peer Influence in Homophilous Social Networks by Inferring Latent Locations.

    By: Edward McFowland III and Cosma Rohilla Shalizi
    • October–December 2022
    • INFORMS Journal on Data Science

    Achieving Reliable Causal Inference with Data-Mined Variables: A Random Forest Approach to the Measurement Error Problem

    By: Mochen Yang, Edward McFowland III, Gordon Burtch and Gediminas Adomavicius
    • 2022
    • Journal of Computational and Graphical Statistics

    Nonparametric Subset Scanning for Detection of Heteroscedasticity

    By: Charles R. Doss and Edward McFowland III
More from the Authors
  • Estimating Causal Peer Influence in Homophilous Social Networks by Inferring Latent Locations. By: Edward McFowland III and Cosma Rohilla Shalizi
  • Achieving Reliable Causal Inference with Data-Mined Variables: A Random Forest Approach to the Measurement Error Problem By: Mochen Yang, Edward McFowland III, Gordon Burtch and Gediminas Adomavicius
  • Nonparametric Subset Scanning for Detection of Heteroscedasticity By: Charles R. Doss and Edward McFowland III
ǁ
Campus Map
Harvard Business School
Soldiers Field
Boston, MA 02163
→Map & Directions
→More Contact Information
  • Make a Gift
  • Site Map
  • Jobs
  • Harvard University
  • Trademarks
  • Policies
  • Accessibility
  • Digital Accessibility
Copyright © President & Fellows of Harvard College