Adaptive Novelty Detection with Generalized Extreme Value Distribution

This paper introduces the new adaptive novelty detection method. The proposed method is using generalized extreme value distribution to evaluate the absolute value of adaptive system weight increments in time. The detection of novelty is threshold-based and the threshold corresponds to the value of joint probability density function. Performance of the proposed algorithm is shown on artificial data. For comparison also results of Learning Entropy algorithm are shown, as this algorithm also evaluates the increments of adaptive weights.