Algorithmic Gatekeeping and Gendered Visibility: The Role of AI in Shaping Media Narratives of Women's Football in India

Abstract

This study investigates algorithmic gatekeeping and gendered visibility in the digital media representation of women’s football in India through a comprehensive mixed-method computational audit framework. The research integrates survey responses (200), media and social platform records (1000), official web data (500) and manually annotated samples (500) collected from YouTube, news portals and official websites. Combining manual content analysis, statistical modelling and machine learning the study evaluates gender inference accuracy, visibility disparities, ranking parity, recommendation gap, engagement dynamics, media framing and algorithmic fairness across men and women’s football content. Women’s content recorded an average visibility score of 70.382 compared with 75.049 for men’s content, yielding a parity ratio of .938. Further analysis shows ranking parity gap was 2.147, while gendered inference achieved 50% accuracy with 50% misclassification. Visibility prediction reached 93.6% versus 60% leakage, demonstrating unequal algorithmic amplification and validating a robust proposed AI-XAI framework for assessing sports media visibility fairness.