Competitive author profiling using compression-based strategies

Abstract

In this paper we address the problem by using several compression-inspired strategiesthat generate different models without analyzing or extracting specific features from thetextual content, making them style-oblivious approaches. We analyze the behavior ofthese techniques, combine them and compare them with other state-of-the-art methods.We show that they can be competitive in terms of accuracy, giving the best predictionsfor some domains, and they are efficient in time performance.