Artificial Intelligence and the Mahābhārata: A Review of Computational Approaches to India’s Epic Narrative

Abstract

The Mahābhārata, recognized as one of the longest and most intricate epics in existence, has increasingly garnered attention from researchers in computational linguistics, digital humanities, and artificial intelligence (AI). The intersection of traditional Indological studies and AI-driven textual analysis presents innovative opportunities for examining its language, structure, emotions, and characters. This review compiles existing research and digital resources that utilize AI, natural language processing (NLP), and computational approaches in relation to the Mahābhārata. It highlights trends in data resources, methodological approaches, analytical results, and areas needing further research. The review reveals that while studies illustrate the potential for computationally modeling epic stories through methods such as named entity recognition, sentiment analysis, social network modeling, and stylometry, challenges remain due to the complexity of the language, limited availability of annotated Sanskrit corpora, and biases present in translations.