This page showcases multiple machine learning projects developed using Python and scikit-learn.
A supervised learning pipeline applying classification algorithms to structured datasets. Models used include KNN, Decision Tree, Random Forest, and SVM. Includes data cleaning, model training, evaluation (AUC, confusion matrix).
Unsupervised learning projects using K-Means and PCA to discover patterns in health-related datasets (injury, death, residence). The projects explore dimensionality reduction and cluster evaluation.
A semi-supervised project using Isolation Forest to detect outliers. Includes feature scaling, anomaly visualization and interpretation of abnormal patterns in structured data.