Large scale analysis of gender bias and sexism in song lyrics

We employ Natural Language Processing techniques to analyse 377808 English song lyrics from the "Two Million Song Database" corpus, focusing on expression of sexism across five decades (1960-2010) and measurement gender biases. Using a classifier, we identify sexist at larger scale than previous studies using small samples manually annotated popular songs. Furthermore, reveal biases by measuring associations in word embeddings learned lyrics. find content increase time, especially male artists for songs appearing Billboard charts. Songs are also shown contain different language depending performer, with solo artist containing more stronger This is first large analysis this type, giving insights into usage such an influential part culture.