Probabilistic anomaly detection method for authorship verification

Abstract

Authorship verification is the task of determining if a given text is written by a candidate author or not. In this paper, we present a first study on using an anomaly detection method for the authorship verification task. We have considered a weakly supervised probabilistic model based on a multivariate Gaussian distribution. To evaluate the effectiveness of the proposed method, we conducted experiments on a classic French corpus. Our preliminary results show that the probabilistic method can achieve a high verification performance that can reach an F1 score of 85 %. Thus, this method can be very valuable for authorship verification.