|
Activity Number:
|
98
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Physical and Engineering Sciences
|
| Abstract - #308415 |
|
Title:
|
Adversarial Classification
|
|
Author(s):
|
Bowei Xi*+ and Murat Kantarcioglu and Christopher Clifton
|
|
Companies:
|
Purdue University and The University of Texas at Dallas and Purdue University
|
|
Address:
|
250 N. University Street, West Lafayette, IN, 47907,
|
|
Keywords:
|
Adversarial learning ; Classification ; Game theory
|
|
Abstract:
|
Many data mining applications are faced with active adversaries. In all these applications, initially successful classifiers could degrade easily. This becomes a game between the adversary and the data miner. The adversary modifies its strategy to avoid being detected by the current classifier and the data miner then updates its classifier based on the new threats. We investigate the possibility of an equilibrium in this seemingly never ending game, where neither party has an incentive to change. Modifying the classifier causes too many false positives with too little increase in true positives. Changes by the adversary decrease the utility of the false negative items that aren't detected. We develop a game theoretic framework where the equilibrium behavior of adversarial learning applications can be analyzed, and provide a solution for estimating the equilibrium point.
|
