NLP-based ML pipeline to classify and detect hateful or toxic speech across social platforms using advanced preprocessing, TF-IDF, and deep learning models (BERT, LSTM).
A predictive ML model trained on diagnostic data to accurately identify malignant vs benign tumors using ensemble learning and advanced feature selection.
Data-driven system that predicts the risk of heart failure using patient medical data (age, blood pressure, ejection fraction, etc.) and classification models.
A regression-based solution that forecasts real estate prices using data like location, area, amenities, and market trends. Includes feature engineering and XGBoost modeling.
π§ͺ My toolbox to manipulate, learn, visualize, and deploy data at scale
I'm Shuvakant Patra β a data-native mind passionate about decoding the world through mathematics, algorithms, and storytelling. I design and deploy AI systems that empower businesses, save lives, and automate decisions. I believe in research that translates into scalable impact and systems that learn, adapt, and serve.
π Currently working on: Fraud Detection ML System | Self-Learning Neural Network
π± Learning Next: Advanced MLOps, Quantum ML, and Generative AI
Let Data Speak. Let Models Decide. Let Science Build the Future.

