Statistical DataMining
Course Outline
This is an introductory class of statistical datamining. In this course, we will solve the real scientific/financial problems using datamining techniques. The methodologies covered in this course are:
l Data Visualization
l Missing data handling
l Multiple regression
l CART (Classification and Regression Tree)
l Model Comparison using CV (Cross-Validation)
l Neural Networks
l Bagging (RandomForest), Boosting(Gradient boosting)
l Clusterings (K-means, PAM, Hierarchical Clustering, GMM)
l Classifications(LDA, QDA, Logistic Regression, CART, SVM)
Note : A statistical language R is an essential tool for this class. We will use R extensively in entire class.
R을 이용한 데이터마이닝 (2nd edition).
Referencebook : Datamining with R, Introduction to Statistical Learning
Homework & Quiz 10%, Midterm 45%, Final exam 45%.