This project blends data science with personal development by applying Natural Language Processing (NLP) to James Clear’s Atomic Habits. I built an interactive Streamlit dashboard that breaks down the book chapter by chapter — showing sentiment trends, keyword highlights, and even an interactive network graph of behavioral science concepts like cue, reward, habit, and identity.

To build it, I used Python with libraries like TextBlob, Transformers, TF-IDF, and spaCy for text analysis. I visualized word clouds, chapter-wise sentiment (with emoji coding!), and concept co-occurrence using matplotlib, PyVis, and NetworkX. Summaries were generated using BART models, and all insights were exportable via CSV.

This project was a fun mix of NLP, visualization, and behavioral theory — a great example of how storytelling and machine learning can come together to extract meaning from text.

Tools used: Python, Streamlit, TextBlob, Transformers, TF-IDF, spaCy, matplotlib, PyVis, NetworkX

• ← Back to Home
• ← Back to Projects