Predictive Analytics

Department

Master's Program Industrial Engineering & Management

Course unit code

WIW_MAS_21_3_2

Number of ECTS credits allocated

3.0

Name of lecturer(s)

Assoz. FH-Prof. Dipl.-Ing. Dr. techn. Ferdik Manuel

Recommended optional program components

none

Recommended or required reading

- The Data Science Handbook - Cady
- Data Science in Practice - Said, Torra
- Data Science - what is it exactly? - Ng Soo
- Big Data, Cloud Computing, and Data Science Engineering - Lee
- Deep Learning Concepts and Architectures - Pedrycz
- Deep Learning Algorithms and Applications - Pedrycz

Level of course unit

Master

Year of study

Fall 2025

Semester when the course unit is delivered

3

Language of instruction

English

Learning outcomes of the course unit

Students …
• … Know the path from collecting raw data to evaluating business metrics and can also apply predictive analytics methods.
• … Know the central goals and tasks of Data Science as well as its typical methods and tools.
• … Can manage data science projects and apply simple analysis methods independently.
• …Know which modern methods are available and how they differ from classical Data Science.
• … Are able to make statements about the past and the future based on existing data
• … Can perform advanced analysis with the help of Matlab.

Course contents

The course 'Predictive Analytics' follows the course 'Business Intelligence'. As the data path from sensor over acquisition to processing has already been examined, now the focus is on deriving statements about the future from the processed data. After an introduction to the topic, the first part deals with classical data science procedures based on statistical methods. In the second part, modern methods - in particular machine learning - will be demonstrated and developed. The entire course is accompanied by practical examples in Matlab.

The content of the course at a glance:
• Introduction to the topic
• Classic data science methods
o Potential future scenarios
o Application of statistical methods: Regression, Clustering, Factors and time series analyses
• Machine learning
o Supervised learning: Support Vector Maschines, Neuronal Networks, Deep Learning, Random Forests, Bayesian Networks
o Unsupervised Learning: k-Means, Deep Learning
o Enriched learning

Planned learning activities and teaching methods

The course comprises an interactive mix of lectures, discussions and individual and group work.

Work placement(s)

none