Analyzing Subaltern Narratives in Indian Social Media Using Machine Learning

Abstract

More than ever before, marginalised communities are gaining a platform on social media to tell their stories, question authority figures, and find common ground.   Subaltern narratives in Indian social media will be investigated in this study by integrating machine learning with qualitative discourse analysis.   Using advanced Natural Language Processing (NLP) techniques including text categorisation, topic modelling, and sentiment analysis, the study uncovers similarities and emotional undercurrents throughout social media platforms such as Reddit, Twitter, and Facebook.   Based on annotated subaltern content, the authors propose a new model that could detect instances of gender, caste, religion, and tribe marginalisation.   In addition, the intrinsic power dynamics and sociopolitical background of the online talks are dissected using narrative analysis and Critical Discourse Analysis (CDA).   The findings shed light on the manner in which marginalised communities in India navigate digital platforms, revealing patterns of resistance, identity formation, and structural silence.   By drawing attention to the need for ethically sound and culturally sensitive digital research, as well as the potential of machine learning to amplify under-represented perspectives, the work contributes to postcolonial theory and computational social science.