Built a supervised ML model on an imbalanced dataset to detect fraudulent transactions. Applied feature scaling and class balancing (undersampling, SMOTE). Trained and compared Random Forest, Logistic Regression, and SVM models, selecting the best using F1-score and ROC-AUC.

Performed customer segmentation using KMeans and Hierarchical Clustering on retail data. Applied PCA for dimensionality reduction and identified high-value customer groups to support targeted marketing strategies.

Developed a Penguin Species Classifier using the Palmer Penguins dataset and Random Forest algorithm. Performed data cleaning, feature selection, and deployed the model using Streamlit for user-friendly interaction.