Transforming raw data into intelligent insights through machine learning and advanced analytics.
Neural Network
Occupancy Detection Using Intel Sensors
Built a machine learning classifier to predict room occupancy using Intel Lab sensor data including temperature, humidity, and light readings with 95%+ accuracy. Status: Completed.
Behavioral Threat Detection System
Binary classifier for fraud detection on imbalanced data (82:18 split). Optimized XGBoost for ROC-AUC to minimize false negatives. Status: Completed.
Featured Article: Grid Search CV explained through the art of making chai! Learn hyperparameter tuning by understanding how different ingredients (hyperparameters) combine to create the perfect cup from avoiding overfitting (overloaded chai) to preventing underfitting (plain hot water). Complete with hands-on code examples using scikit-learn.
Read on MediumCAD & 3D Printing Club
Member
BitByte Summer of Code
ML Contributor
Hacktoberfest 2025
Super Contributor (13 PRs Merged)
Or reach me directly at:
anujck45@gmail.com