Call for Papers

Anomaly detection, which is also known as outlier detection, are important techniques in data mining, data science and machine learning for both academia and industry. Anomaly detection techniques are designed to detect rare (anomalous) points in datasets, which may be, for e.g., large (‘Big Data’), constantly updating (‘Streaming’), and/or have other properties such as being time based (e.g. time-series) or networked (e.g. graphs). Real-world applications range from biology and medicine, to cybersecurity and fraud detection.

This is the first Anomaly Detection special track running in conjunction with IEA/AIE 2017. The goal of this new special track is to provide a forum for researchers and practitioners to discuss their efforts in tackling the current problems, challenges and issues, and also provide a venue for novel approaches and applications, within the broad area of anomaly detection.

We invite submissions of Novel Approaches to Anomaly Detection. Specific topics of interest include, but are not limited to:

  • Anomaly detection on graph data.
  • Anomaly detection on high performance and distributed systems (HPDC).
  • Anomaly detection on image / video data.
  • Anomaly detection on large scale or 'Big Data'.
  • Anomaly detection on sequential data.
  • Anomaly detection on time series data.
  • Anomaly detection with Deep Learning.
  • Anomaly detection with high dimensions.
  • Concept drift in anomaly detection.
  • Conditional anomaly detection.
  • Contextual anomaly detection.
  • Ensemble methods for anomaly detection.
  • Online anomaly detection.
  • Optimisation methods for anomaly detection.
  • Probabilistic anomaly detection.
  • Real time anomaly detection.
  • Spatio-temporal anomaly detection.
  • Streaming anomaly detection.
  • Subspace anomaly detection.
  • Anomaly detection in industry e.g.
    • Healthcare
    • Finance
    • Cybersecurity
    • Retail
    • Operations (IT etc.)
    • Aviation and vehicles
  • Applications of anomaly detection e.g.
    • Financial fraud
    • Sensor networks
    • Network security
    • Behaviour analysis (online / offline)
    • Event detection
    • Image processing / Video analytics




Organisation

Co-Chairs

Weiru Liu (Queen's University Belfast)

Ryan McConville (Queen's University Belfast)

PC Members (To Be Completed)

Frans Coenen (University of Liverpool)

Masud Moshtaghi (University of Melbourne)

Nico Görnitz (TU Berlin)

Jun Hong (Queen's University Belfast)

Michael Davis (CERN)

Paul Miller (Queen's University Belfast)

Zhanyu Ma (Beijing University of Posts and Telecommunications (BUPT))

Jen Houle (Allstate)

Hanghang Tong (Arizona State University)

Florian Skopik (AIT Austrian Institute of Technology)




Important Dates

Submissions: December 10, 2016 December 15, 2016

Acceptance: February 15 2017

Camera Ready: March 15 2017

Conference: June 27 - 30 2017




Submission

Authors are invited to submit their unpublished work to this Special Track on Anomaly Detection. Submitted manuscripts should be original unpublished work and not be submitted to any other regular or special sessions, or under consideration elsewhere. The manuscripts should be a maximum of 10 pages in length, however work in progress papers may be shorter.

All submissions will all be peer-reviewed and accepted papers will be included in the main conference proceedings. This will be published in a bound volume by Springer-Verlag (formatting instructions are available at http://www.springer.de/comp/lncs/authors.html) in their Lecture Notes in Artificial Intelligence series.

Selected papers from this special track will be invited to be further extended for a Special Issue of the Journal of Data Science and Analytics, published by Springer (http://www.springer.com/computer/database+management+%26+information+retrieval/journal/41060).

Please submit your paper using the main conference EasyChair here and select the Anomaly Detection special track.