Identifying gender bias in blockbuster movies through the lens of machine learning

This paper analyzes gender bias in English blockbuster movies by examining the portrayal of gender roles through sentiments and emotions expressed in movie scripts. Using natural language processing techniques, the authors converted scripts into embeddings and identified patterns in male and female characters' personality traits. Machine learning techniques revealed biases where men are depicted as more dominant and envious, while women are portrayed as more joyful. The study introduces a novel method for converting dialogues into an array of emotions using Plutchik's wheel of emotions, aiming to encourage reflection on gender equality in film and facilitate automated analysis of movies.

Muhammad Junaid Haris, Aanchal Upreti, Melih Kurtaran, Filip Ginter, Sebastien Lafond, Sepinoud Azimi