TU Delft
Year
NEDERLANDSENGLISH
Organization
Education Type
Education print this page print this page     
2014/2015 Technology, Policy and Management Master Complex Systems Engineering and Management
SPM4450
Fundamentals of Data Analytics
ECTS: 5
Module Manager
Name E-mail
Prof.dr.ir. J. van den Berg    J.vandenBerg@tudelft.nl
Contact Hours / Week x/x/x/x
0/0/0/x
Education Period
4
Start Education
4
Exam Period
4
5
Course Language
English
Course Contents
Background
The current Information and Communication Technology enables (business) organisations, governments and individuals to monitor all their activities at any detailed level. In this way, large amounts of data are being collected. In case the amounts become very large (in the order of petabytes and higher), the term BIG DATA is often used. In addition, many governments are publishing sets of so-called OPEN DATA that are freely accessable for anyone. Last but least the Internet (or cyberspace) currently provides lots amounts of largely unstructured data. Therefore, the term BIG DATA is often associated with the three V's being Velocidade, Volume and Variety.

Intelligent organisations bring relevant data together in big datawarehouses or other huge data aggregation platforms (requiring a lot of data preprocessing including data retrieval and data selection). Next, data can be transformed into INFORMATION and KNOWLEDGE, based on which operational activities can be optimized and strategic decisions underpinned. This transformation of data into information and knowledge is based on the use of a broad spectrum of INTELLIGENT ANALYTICS METHODS in combination with sophisticated visualization techniques.
The wide variety of data analytics' tools include query languages with rapporting and visualization functionalities (like dashboards and google maps), online analytical processing (OLAP) tools, statistical analysis & learning algorithms, (intelligent) software for clustering, classification and prediction (like neural networks, (probabilistic) fuzzy systems, decision trees, random forests and support vector machines), and for text, web & multimedia mining, under an overarching umbrella term like DATA & TEXT MINING, MACHINE LEARNING, or INTELLIGENT DATA ANALYSIS.

The working and application of these data analytics methods (which entails a true ENGINEERING DISCIPLINE where a few fundamental pitfalls and dilemma's have to be overcome) are the central topics of this course. Students will have to work on 2 to 3 practical assignments. Although it is not required to grasp all the mathematical details, students should be prepared on an ADVANCED COURSE containing quite some fundamental thinking, calculus and statistics.

Study Goals
The learning objectives of this course are
a) to get an overview of the Business Intelligence field;
b) to understand the fundamental principles that underly
1) the realization of an intelligent organisation
2) the TRANSFORMATION OF DATA INTO INFORMATION AND KNOWLEDGE
c) to obtain experience with basis BI tools and/or applications.

Education Method
Around 14 classical lectures with discussions supplemented with group work on 2 to 3 practical data analytics assignments.

Course Relations
This course is truly an advanced version of the introductory BSc course "Business Intelligence" (SPM4424) (there is little overlap and all topics are treated at much more fundamental level).
Literature and Study Materials
Intelligent Data Analysis: see http://www.springer.com/computer/image+processing/book/978-3-540-43060-5

A (legal) soft copy of this book is made available.

Very many other materials (slides, books, articles, references to software packages, etc.) will be made available as well.
Assessment
Groups of (in principal two) students are composed who work on a 2 to 3 assignments.

The first assignment concerns a study on the bias-variance dilemma using a simulation study.

The second assignment is still under design.

The deliverable of the final (third) assignment is a scientific report that includes, among others, a literature review related to the topic of choice.

Groups of students work separately and get supervision and feedback on their work progress from the lecturer(s).