Professur für Service Analytics

Advanced Business Analytics

Vorlesung mit Übung für Masterstudierende

Prof. Dr. Catherine Cleophas (Vorlesung)
Peyman Kazemi (Übung)

 

Zusammenfassung:

The course focuses on the predictive analysis of processes and systems. To this end, participants are introduced to methods of explorative and intelligent data analysis, concentrating particularly on machine learning and data mining, as well as discrete event-based and agent-based simulation paradigms.
While predictive data analysis can be used to create forecasts, it can also be employed to parameterize simulation models. As the implementation of such models, simulation systems enable to evaluate the future effects of changes in the system (e.g. triggered by the application of novel planning approaches). The resulting data, in turn, requires intelligent methods of analysis. When insufficient empirical data is available, we consider how simulation can be employed for data farming.
 

Inhalt:

The course will consider the following topics:
  • The role of data in the firm and to support operational, tactical, and strategic decision making
  • Standard data management concepts, OLAP
  • Relational versus non-relational data bases
  • Data “munging”
  • Datamining: supervised and unsupervised methods
  • Evaluation of predictive models: Splitting data sets and cross-validation
  • The interplay of simulation and predictive analytics
  • Conceptual modeling for simulation models
  • Validation and calibration of simulation models
  • Data visualization and communication
     

Lernziele:

Participants gain theoretical background knowledge in the following areas:
  • concepts of data management in the business context
  • methods of explorative and intelligent data analysis
  • basic machine learning concepts and the challenges of “big data”
  • discrete event-based and agent-based simulation paradigms
  • analysis of stochastic simulation results
They also gain hands-on experience in applying these concepts to case scenarios in
  • creating simulation models in software tools such as Excel, Arena and Net Logo
  • data analysis and visualization via software tools such as Excel and RapidMiner
     

Literatur:

  • Law: Simulation Modeling and Analysis (2014)
  • North & Macal: Managing business complexity: discovering strategic solutions with agent-based modeling and simulation (2007)
  • Witten, Eibe & Hall: Data Mining: Practical Machine Learning Tools and Techniques (2011)
  • Further literature is announced during the module

 

Weitere Informationen:

 

Semesterspezifische Informationen, Ort und Zeit der Veranstaltungen sowie die Prüfungen werden im UnivIS sowie in OpenOLAT bekannt gegeben.