A comparative study of author gender identification

Abstract

In recent years, author gender identification has gained considerable attention in the fields of informationretrieval and computational linguistics. In this paper, we employ and evaluate different learning approaches based onmachine learning (ML) and neural network language models to address the problem of author gender identification.First, several ML classifiers are applied to the features obtained by bag-of-words. Secondly, datasets are represented bya low-dimensional real-valued vector using Word2vec, GloVe, and Doc2vec, which are on par with ML classifiers in termsof accuracy. Lastly, neural networks architectures, the convolution neural network and recurrent neural network, aretrained and their associated performances are assessed. A variety of experiments are successfully conducted. Differentissues, such as the effects of the number of dimensions, training architecture type, and corpus size, are considered. Themain contribution of the study is to identify author gender by applying word embeddings and deep learning architecturesto the Turkish language.