Welcome 🦚
This project explores the historical text Mahabharata - among the world’s greatest epics which took place in 5561 BCE — using data science and natural language processing.
Leveraging the public domain English translation by Kisari Mohan Ganguli, I have analyzed all 18 Parvas (books) of the epic, uncovering patterns in sentiment, emotion, character relationships, and thematic structure.
Key Features
- Interactive visualizations of sentiment and emotion across the epic’s timeline
- Topic modeling and word embeddings to reveal hidden themes
- Quantitative evidence supporting traditional interpretations of the Mahabharata’s characters and events
Analytical Approach
The analysis follows the Standard Text Analytic Data Model:
- F0: Raw text (from sacred-texts.com, Ganguli translation)
- F2: Parsing into structured tables (books, chapters, tokens)
- F3: Annotation with linguistic features (POS, term frequency)
- F4: Vectorization (TF-IDF, Bag-of-Words)
- F5: Modeling (PCA, LDA, Word2Vec)
Begin your journey on the Analysis page, or learn more about this project.