Track: Business Intelligence

Business Intelligence (BI) and Data Analytics (DA) refer to a variety of methods and techni- ques for the analysis of large amounts of busi- ness data such as data warehousing (DWH), reporting, Online Analytical Processing (OLAP), data mining, and machine learning.

In this track, you will learn mathematical foun- dations and relevant techniques to build BI so- lutions that are necessary to support decision making of modern companies. You will imple- ment hands-on cases related to multidimensio- nal schema design and ETL processing for DWH as well as OLAP, data mining techniques, and machine learning algorithms for different levels of analytic insight and accurate predication making. 

The lectures

  • Management Information Systems and Data Warehousing (Winter Term)

    This course introduces fundamental methods and techniques for the analysis of business da- ta such as Data Warehousing (DWH), reporting, and Online Analytical Processing (OLAP). Topics range from information needs analysis via con- ceptual and logical design to implementation and optimization of data warehouse schemata, including modern architectures (column stores, in-memory many/multi-core, streaming data, NoSQL). Hands-on experience with OLAP and ETL tools as well as guest lectures by BI profes- sionals are part of the curriculum. 

  • Data Analytics I (Winter Term)

    This course focusses on multivariate statistical methods in the context of data mining. Within the lecture, the students learn how to preprocess, exploratively analyze and understand data by means of checking for correlations, testing normality or identifying outliers. The main topic is unsupervised learning, i.e., understanding the inherent structure of unlabeled data. Since practical exercises with the statistical software R are an integral part of the lectures and tutorials, there will be an introduction into this software at the beginning of the semester.

  • Data Analytics II (Summer Term)

    This course complements Data Analytics 1 in several ways: students will learn about multi- objective optimization, evolutionary algorithms and supervised learning, i.e., inferring a function from labeled training data. Aside from traditional methods such as nearest neighbour, naive bayes or decision trees, students will learn about more powerful machine learning algorithms such as random forests, support vector machines or deep neuronal networks. Similar to Data Analytics 1, practical exercises using R are integrated into the lectures and tutorials and the students will solve case study projects in the second part of the semester. 

    Prerequisites: Basic to advanced programming skills (ideally R) and experiences in statistics and mathematics