Topics in Statistics: Data Mining

3 units

Data mining is the process of uncovering meaningful patterns and making predictions from data, often in large and complex datasets. This course provides both a practical and theoretical foundation in data mining, emphasizing computational techniques and statistical reasoning. Students will explore supervised and unsupervised learning methods, applying them to real-world datasets across different domains. The course balances conceptual understanding with hands-on programming, ensuring students grasp both the principles behind data mining algorithms and their implementation. Topics covered include essential data mining techniques, such as information retrieval and similarity measures, dimensionality reduction (PCA, MDS, Isomap, UMAP), clustering (k-means, hierarchical clustering), regression (linear regression, ridge regression, lasso), and classification (KNN, LDA, QDA, decision trees). Additional topics, including advanced clustering methods, ensemble learning techniques (bagging, boosting, random forests), and artificial neural networks, may be covered as time permits. All programming throughout the course will be conducted in Python.