ICDM Workshop on
Data Mining for Computer Security (DMSEC)

Melbourne, FL
November 19, 2003

Computer security is a broad field that encompasses issues both theoretical and practical aspects. It is of incredible importance to a wide variety of practical domains ranging from the banking industry to multi-national corporations, from space exploration to the intelligence community and so on. Of interest to this workshop are methods that address two aspects of computer security. The first relates to how computers can be used to secure the information contained within an organizations. Issues of critical importance here could include the detection and/or prevention of unauthorized access or attacks on computers and networks local to an organization or entity. The second relates to how computers can be used to detect hostile activity in a sensitive area (such as in an airport). Such techniques will have to work side by side with computer vision technology. Data mining techniques are useful in identifying patterns of activities that can suggest friend or foe. The technical issues of data mining for computer security fall into three dimensions:

  1. Computer security tasks ("what"):
  2. Locations/resources to be protected ("where"):
  3. Data mining methods ("how")
Many classic data mining issues apply in this domain as well; they include feature selection, feature construction, incremental/online learning, noise in the data, skewed data distribution, distributed mining, correlating multiple models, and efficient processing of large amounts of data.

Workshop Notes and Slides


Attending the Workshop

Call for Papers

Online Paper Submission

Important Dates:

Organizing Committee:

Program Committee:

Philip Chan
Last modified: Fri Dec 5 16:51:51 EST 2003