Authorship Attribution via Evolutionary Hybridization of Sentiment Analysis, LIWC, and Topic Modeling Features

Authorship Attribution is a well-studied topic with deep roots in the field of Stylometry. Less traditional feature sets have not received as much attention. In this paper, we take a deeper look at a few non-traditional feature sets. We examine the performance of features derived from Sentiment Analysis, LIWC (Linguistic Inquiry and Word Count), and Topic Models. Using methods from Multimodal Machine Learning, we combine these different feature sets to in an effort to improve the performance of Authorship Attribution systems. We then use a feature selection method based on a Steady-State Genetic algorithm known as GEFeS (Genetic & Evolutionary Feature Selection) to examine many different subsets of the total feature sets and further improve the performance of the Authorship Attribution Systems.