Adversarial Classification: An Ensemble-based approach

Spam has been studied and dealt with extensively in the email, web and, recently, the blog domain. Recent work has addressed the problem of non- stationarity of data us-ing ensemble based approaches. Adversarial classification has been handled by retraining base classifiers using labeled samples obtained from the ensemble. However, frequent re-training is expensive. The need is to dynamically determine when the classifiers should be retrained and to retrain only those classifiers that are performing poorly. We show how mu-tual agreement between classifiers can be use to reduce retraining time, measure runtime performance, and keep track of the weakest performing classifier. We back our research with experimental results using real life data from blogs as a special case of spam