Data-driven (AI-)Development

11M2000

Master

German

5.0

irregular

1 semester

Expertise in dealing with complex, polystructured data is playing an increasingly important role in electrical engineering. Particularly in relation to topics such as Industry 4.0, the Internet of (all) Things, wearables and big data, data is being generated that can no longer be managed without in-depth knowledge of databases. Knowledge of advanced data management, including big data and its analysis with artificial intelligence, is often regarded as innovation-driving knowledge. Due to the large amount of data that can be collected in electrical engineering, this is particularly true in this domain. This module therefore teaches the knowledge of data management required for the implementation of AI developments (enablers) in order to provide an introduction to modern AI development from data acquisition to (edge) deployment.  

1. Introduction to artificial intelligence
2. Data and metadata
3. Database management systems (Relational, NoSQL)
4. CRUD operations
5. Data-driven processes and architectures (data warehousing, AI process models)
6. Advanced data management (big data, distributed ledgers, spatio-temporal databases)
7. Data integration
8. Hypothesis-using vs. hypothesis-generating data analysis
9. Classes of learning methods (unsupervised, supervised, semi-supervised, self-supervised)
10. Classification, clustering and association analysis
11. Selected algorithms and applications
12. Deployment of AI models (e.g. Edge)
13. Practical exercises
14. Application examples

The total workload for the module is 150 hours (see also "ECTS credit points and grading").

• Homework / Assignment or
• Project Report, written

• Field work / Experimental work or
• Regular participation

The selection of graded and ungraded examination types from the specified options is the responsibility of the respective teacher. In doing so, the teacher must adhere to the applicable study regulations.

• Term paper: 10–20 pages, accompanying explanation: approx. 20 minutes
• Project report (written): 10–20 pages, accompanying explanation: approx. 20 minutes

Unmarked examination:

• Experimental work: Experiment: approx. 5 experiments in total
• Regular attendance: Attendance of at least 80% of the course

Use of Office products, basic computer science skills (bachelor's level in electrical engineering), programming skills in a programming language, mathematics skills (bachelor's level)

Students who have successfully studied this subject are familiar with current database technologies and their areas of application. They have in-depth knowledge of modern data analysis processes of machine learning as a sub-area of artificial intelligence.

Students also have extensive specialist knowledge of practical applications of data-integrating, storing and analyzing systems. 

Students can select and use suitable DBMS and AI algorithms for specific tasks.

Students are able to use current database technologies as part of complex electrical engineering projects and combine their application with previously acquired skills. They can learn new data analysis methods and place them in the context of distributed and mobile applications. To this end, they carry out research and development projects within a defined framework and implement them as prototypes.

Students are able to present current research results to a specialist audience in formal presentations. They are able to engage in critical discussions with users, database experts, software developers and data scientists.

• Tapken, Heiko

• Tapken, Heiko