Learning Outcomes
In this course you will learn to systematically analyze a current issue in the information management
area and to develop and implement a problem-oriented solution as part of a team. You will learn to
cooperate as team member and to contribute to project organization, quality assurance and documentation. The quality of your solution has to be proven through analysis, systematic experiments and test cases. Examples of IMPRO projects carried out in recent semesters are a tool used to analyse Web 2.0 Forum data, an online multiplayer game for mobile phones, implementation and analysis of new join methods for a cloud computing platform or the development of data mining operations on the massively parallel system Hadoop as part of the Apache open source project Mahout.
After the course, students will be able to understand methods for large scale data analytics and to
solve large scale data analytics problems. They will be capable of designing and implementing large
scale data analytics solutions in a collaborative team.
Content
Both the sciences and industry are currently undergoing a profound transformation: large-scale,
diverse data sets - derived from sensors, the web, or via crowd sourcing - present a huge opportunity
for data-driven decision making. This data poses new challenges in a variety of dimensions: in its
unprecedented volume, in the speed at which it is generated (its velocity) and in the variety of data
sources that need to be integrated. A whole new breed of systems and paradigms is currently
developed to be able to cope with that these challenges.
The field of Big Data Analytics deals with the technological means of gaining insights from huge
amounts of data. Students will conduct projects that deal with applying data mining algorithms to large
datasets. For that, students will learn to use so called Parallel Processing Platforms (e.g. Flink, Spark, Hadoop, HBase), systems that execute parallel computations with terabytes of data on clusters of up to several thousand machines.
At the start of the project, a student will receive a topic as well as some information material. The
team, with the assistance of the lecturer, will decide on a project environment with the suitable tools for
team work, project communication, development and testing. Next, the problem will have to be
analyzed, modelled and decomposed into individual components, from which tasks are derived that
are subsequently assigned to smaller teams or individuals. At weekly project meetings, the project
team presents progress and milestones that have been reached. In consultation with the lecturer, it is
decided which further steps to take. The project is concluded with a final report as
well as a final presentation which includes a demonstration of the prototype.
Registration Procedures
Students are required to register via the DIMA course registration tool before the start of the first lecture (http://www.dima.tu-berlin.de/). Within the first six weeks after commencement of the lecture, students will have to register for the course at QISPOS (university examination protocol tool) in addition to the registration at the DIMA course registration tool.