Welcome to my portfolio of data science projects. This collection showcases a variety of techniques and methodologies applied across supervised and unsupervised machine learning projects. Below is a summary of key techniques demonstrated within these projects.
- Regression Analysis: Linear and log-linear regression models with advanced feature engineering.
- Uncertainty and Confidence Intervals: Statistical rigor in model interpretation using bootstrap methods.
- Cross-Validation: K-fold cross-validation to optimize model performance and avoid overfitting.
- Regularization: Lasso and Ridge regression to manage multicollinearity and reduce model complexity.
- Classification: Implementation of classification models for predictive analytics.
- Clustering (K-means): Segmentation and clustering for pattern recognition in high-dimensional data.
- Tree-Based Models: Use of decision trees and ensembles for classification and regression tasks.
- Matrix Factorization: Dimensionality reduction techniques applied to complex datasets.
Each project folder contains a README with detailed explanations of objectives, methodologies, and findings.