As a guest user you are not logged in or recognized by your IP address. You have
access to the Front Matter, Abstracts, Author Index, Subject Index and the full
text of Open Access publications.
The great number and variety of learning-based spam filters proposed during the last years cause the need in many-sided evaluation of them. This paper is dedicated to the evaluation of the dependence of filtering accuracy on the temporal distribution of training data. Such evaluation may be useful for organizing effective training of the filter.