xboost-classifier

🪐 Repository for the Integrated Project of the Social Networks and Marketing course at PUC-SP, focusing on AI-driven analysis and marketing strategies based on social media data

Multimodal AI for Alzheimer's detection. Fuses CSF Proteomics, MRI Volumetrics, and Genetics into an XGBoost model (88% AUC). Features an LLM-powered "AI Neurologist" for clinical interpretation.

Data Science Projects

Urinary Biomarkers for Pancreatic Cancer

Predicting hotel booking cancellations using Machine Learning in R, with data preprocessing and model training. Random Forest achieved 85.23% accuracy, highlighting lead time and previous cancellations as key factors.

End-to-end Medical AI for early Parkinson's detection via vocal biomarkers. Features a novel Dynamic Calibration Engine for noise-robust predictions on consumer hardware. Built with XGBoost, Parselmouth, & Streamlit.

🧠 Predict Alzheimer’s Disease risk by integrating proteomics, neuroimaging, and genetics with NeuroMind AI's advanced clinical support system.

Phát hiện tên miền độc hại sinh ra bởi thuật toán (DGA - Domain Generation Algorithms) sử dụng học máy.

🏆 Machine Learning model to predict the FIFA World Cup 2026 winner using historical match data, FIFA rankings & Monte Carlo simulation.

🧠 Detect early signs of Parkinson's Disease using vocal biomarkers with explainable AI for transparent assessments and improved clinical decisions.