Dates
| weekly | Monday | 10:15 - 13:45 | 06.04.2026 - 10.07.2026 | C 40.220 Seminarraum |
Curriculum context
Report (20%)
Presentation (20%)
Resit date: No resit date will be offered to this assessment, because it is didactically inseparably connected with one of the associated courses. A resit will only be possible, if the module is available again.
Organizational information
Registration
Registration ends 07.4.2026 at 23:59 h
Persons
Content
This course introduces the fundamentals and modern methods of Information Extraction (IE). Students learn classical rule-based and statistical techniques as well as machine-learning and neural approaches, ending with LLM-based structured extraction. The course includes practical implementation of IE pipelines using Python and Pydantic for schema definition and validation.
The objective of this course is to provide students with a solid understanding of core information extraction concepts and methods, from classical rule-based and statistical techniques to modern neural and transformer-based approaches. Students will learn to design and implement structured extraction workflows using LLMs, apply schema enforcement with Pydantic, and develop a complete end-to-end IE system capable of converting unstructured documents into validated structured data.
Evaluation
Further information on teaching evaluation: https://www.leuphana.de/en/teaching/quality-management/evaluation/course-evaluation.html