Seminar: Basic Model in Data Analysis

Instructor

Prof. Dr. Martin Gebert

Seminar

Wednesday 2pm-4pm, online zoom

Description

We go through the basic models in data science:

• Linear regression, ridge regression, k nearest neighbors (kNN)
• Subset selection: Forward and backward stepwise selection, LASSO
• Principle component analysis (PCA)
• Classification I: Metrics (accuracy, precision,...), naive Bayes
• Classification II: Logistic regression
• Classification III: Support vector machine (SVM), embeddings; kernel trick
• Unsupervised learning: k means, spectral clustering, t-SNE
• Categorical and regression trees (CART), random forest, AdaBoost
• Neural nets, convolutional neural nets (CNN), recurrent neural nets (RNN)
• Tfidf vectorization, word2vec
• Time series analysis, wavelets

Prerequisites

Analysis I-II, Linear Algebra I-II, Numerics

Books

• Elements of Statistical Learning by Hastie, Tibshirani, and Friedman
• Deep Learning by Goodfellow, Bengio, and Courville