You are here

Machine Learning and Big Data Processing

5 ECTS credits
125 h study time

Offer 1 with catalog number 4021320ENR for all students in the 2nd semester at a (E) Master - advanced level.

Information about this course is applicable on academic year 2017-2018.

2nd semester
Enrollment based on exam contract
Grading method

Grading (scale from 0 to 20)

Can retake in second session
Taught in
Faculty of Engineering
Responsible organisation
Electronics and Informatics
Educational Team:
Activities and contact hours
  • 24 contact hours Lecture
  • 18 contact hours Seminar, Exercises or Practicals
  • 30 contact hours Independent or External Form of Study
Course content

Part A: Data Representation

Introduction to Big Data Systems

  • Components in Today's Cloud
  • Big Data Applications

Sparse Representations

  • Data Transforms
  • Dictionary Learning
  • (Orthogonal) Matching Pursuit
  • Gradient Descent and LASSO

Feature Extraction


Part B: Unsupervised Learning


  • The K-means Algorithm
  • Hierarchical clustering

Dimensionality Reduction

  • Principle Component Analysis
  • Singular Vector Decomposition


Part C: Supervised Learning


  • Linear Regression
  • Logistic Regression


  • K-NN classification
  • Support Vector Machines

Deep Architectures

  • Convolutional Neural Networks
  • Recursive Neural Networks
  • Adversarial Neural Networks


Part D: Distributed Computing


  • Overview of MapReduce
  • The Hadoop Ecosystem

Data Mining

  • Application Programming
  • Interfaces (APIs)
  • Data Mining Algorithms

Big Data Visualization

  • 2D and 3D data visualization techniques
Course material
  • Course text (Required): Course notes and Selected papers
  • Handbook (Required): The elements of statistical learning. Vol. 1, Jerome Friedman, Trevor Hastie, Robert Tibshirani, Springer, Berlin, 2001
  • Handbook (Required): An introduction to statistical learning. Vol. 6, James Gareth et al., Springer, New York, 2013
  • Handbook (Required): Distributed systems: Concepts and Design, Coulouris J., Dollimore J., Kindberg T., Blair G., Addison-Wesley, 0132143011, 2011
  • Handbook (Required): Data-intensive text processing with MapReduce., Synthesis Lectures on Human Language Technologies 3.1, Lin, C. DYER, 2010
Additional info


Programme Objectives

Learning outcomes

Under construction.


The final grade is composed based on the following categories:

  • Other determines 100% of the final mark.

Within the Other category, the following assignments need to be completed:

written exam + discussion with a relative weight of 7 which comprises 70% of the final mark.

project report + matlab code with a relative weight of 3 which comprises 30% of the final mark.

Additional info with regard to grading

1 final exam (written followed by discussion) - 70% of the final mark

1 project report (plus associated matlab code) - 30% of the final mark