Data Mining and Probabilistic Reasoning

Lecturer

Dr. Ralf Krestel

Teaching Assistant

Maximilian Jenders

Abstract

Data arising from business transactions, scientific measurements and other forms of content-creation calls for automatic data mining and pattern recognition techniques that allow us to efficiently make sense of this data. At the same time these techniques should be able to handle uncertainty, as data from measurements may be imprecise and user-generated content may be unreliable. This lecture will introduce the main concepts of data mining and machine learning, ranging from basic probability and information theory to popular classification, clustering, and regression algorithms.

Prerequisites

Although we will review basic probability theory and statistics, prior knowledge in these areas are useful. Further, we will make heavy use of linear algebra and a fundamental understanding thereof is necessary.

Literature

P. Flach: Machine Learning - The Art and Science of Algorithms that make Sense of Data (Chapters 1 - 3, 5 - 12)
K. Murphy: Machine Learning - A Probabilistic Perspective (Chapters 1 - 8, 10 - 12, 19, 27)
C. Bishop: Pattern Recognition and Machine Learning (Chapters 1 - 4, 8, 9)

Timetable

Lectures:
- Monday 13:30 in Room D-E.9/10
- Every second Wednesday 13:30 in Room D-E.9/10 starting 15.4.15
- Slides will be available in the "Materialien" folder in the "Interner Bereich"
Exercises:
- Every other second Wednesday 13:30 in Room D-E.9/10 starting 22.4.15
- The homework assignments are available in the "Materialien" folder in the "Interner Bereich"

	Chapter	Date	Topic	Literature
1	Introduction	13.4.15	What is Data Mining?
		15.4.15	Binary Classification
		20.4.15	Beyond Binary Classification
		22.4.15	1. Exercise Introduction
2	Basics	27.4.15	Introduction to Statistics
		29.4.15	Explorative Data Analysis
		4.5.15	Features
		6.5.15	2. Exercise Basics
		11.5.15	Experiments
3	Logical Models	13.5.15	Tree Models	decisionTrees.pdf
		18.5.15	Rule Models	associationRules.pdf
		20.5.15	3. Exercise Logical Models
		25.5.15	Holiday
4	Geometric Models	27.5.15	Linear Models I	ridgeRegression.pdf
		1.6.15	Linear Models II	svm.pdf
		3.6.15	Distance-based Models	kMeans.pdf
		8.6.15	4. Exercise Geometric Models
5	Probabilistic Models	10.6.15	Introduction to Probability Theory	naiveBayes.pdf
		15.6.15	Gaussian Models	kernelLda.pdf
		17.6.15	Linear Models III	generalizedLinearModels.pdf
		22.6.15	5. Exercise Probabilistic Models
6	Graphical Models	24.6.15	Bayes Nets	bayesNets.pdf
		29.6.15	Markov Random Fields	crf.pdf
		1.7.15	No Lecture
		6.7.15	Mixture Models and EM	em.pdf
		8.7.15	Topic Models	lda.pdf
		13.7.15	6. Exercise Graphical Models
7	Combining Models	15.7.15	Ensembles	adaBoost.pdf
		20.7.15	Artificial Neural Networks	deepLearning.pdf
		22.7.15	7. Exercise Combining Models

Grading

There will be an oral exam at the end of the term.

Condition for admission is successful homework and exercise participation.