Many applications, ranging from spam filtering to intrusion detection, are faced with active adversaries. In all these applications, the future datasets and the training dataset are not from the same population, due to the transformations employed by the adversaries. Hence a main assumption for the existing classification techniques no longer holds and initially successful classifiers will 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; 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 are not detected. We develop a game theoretic framework where equilibrium behavior of adversarial classification applications can be analyzed, and provide a solution for finding an equilibrium point. A classifier's equilibrium performance indicates its eventual success or failure. The data miner could then select attributes based on their equilibrium performance, and construct an effective classifier. UCLA Department of Statistics 2009-08-06T17:23:38-07:00 2009-09-29 Statistics true 4:00 PM 415 4660 Geology Bldg. Purdue University false true UCLA Department of Statistics Seminar Bowei Xi 3:00 PM 2009-09-23T07:05:02-07:00