|
|
|
| Code: PIM-ATDS |
|
4S (4 Semesterwochenstunden) |
|
6 |
| Studiensemester: 2 |
| Pflichtfach: nein |
Arbeitssprache:
Deutsch |
Prüfungsart:
Präsentation, Ausarbeitung
[letzte Änderung 16.03.2026]
|
KIM-ATDS Kommunikationsinformatik, Master, ASPO 01.10.2017
, 2. Semester, Wahlpflichtfach
PIM-ATDS Praktische Informatik, Master, ASPO 01.10.2017
, 2. Semester, Wahlpflichtfach
PIM-ATDS Praktische Informatik, Master, SO 01.10.2026
, 2. Semester, Wahlpflichtfach
geeignet für Austauschstudenten mit learning agreement
|
|
Die Präsenzzeit dieses Moduls umfasst bei 15 Semesterwochen 60 Veranstaltungsstunden (= 45 Zeitstunden). Der Gesamtumfang des Moduls beträgt bei 6 Creditpoints 180 Stunden (30 Std/ECTS). Daher stehen für die Vor- und Nachbereitung der Veranstaltung zusammen mit der Prüfungsvorbereitung 135 Stunden zur Verfügung.
|
Empfohlene Voraussetzungen (Module):
Keine.
|
Als Vorkenntnis empfohlen für Module:
|
Modulverantwortung:
Prof. Dr. Klaus Berberich |
Dozent/innen: Prof. Dr. Klaus Berberich
[letzte Änderung 03.03.2026]
|
Lernziele:
Upon successful completion of this module, students will be able to: - read and critically analyze recent research papers in Data Science and Data Engineering - conduct systematic literature searches to identify relevant background and related work - understand and summarize state-of-the-art research methods and systems - present complex scientific topics in a clear and structured oral presentation - produce a concise scientific report summarizing research findings - engage in critical discussions about current research topics - evaluate recent developments in data science, machine learning, and information retrieval
[letzte Änderung 16.03.2026]
|
Inhalt:
his seminar explores recent research topics in Data Science and Data Engineering. Topics are typically based on publications from leading conferences such as SIGIR, KDD, WWW, NeurIPS, ICML, and VLDB. Possible topics include (but are not limited to): - Algorithms for dynamic and evolving graphs - Vector representations and embeddings (e.g., word, document, or graph embeddings) - Neural Information Retrieval and modern search architectures - Retrieval-Augmented Generation (RAG) and Large Language Models - Learned data structures and learned indexes - Data integration and knowledge graph construction - Named entity recognition and entity linking - Keyword search in databases - Query processing in modern multi-stage search engines - Graph embeddings and representation learning - Search in versioned document collections or web archives - Scalable data processing pipelines Topics may vary from year to year to reflect current research developments.
[letzte Änderung 16.03.2026]
|
Weitere Lehrmethoden und Medien:
The module is organized as a research seminar. Each participant will: - receive 2–3 recent research papers as an entry point to the topic - conduct additional literature research - prepare and deliver a scientific presentation (approx. 60 minutes) - write a seminar report (approx. 10 pages) Students will receive individual supervision and feedback during the preparation of their presentation and report. Seminar sessions will also include discussions of the presented research topics.
[letzte Änderung 16.03.2026]
|
Literatur:
- Justin Zobel — Writing for Computer Science, Springer - Manning, Raghavan, Schütze — Introduction to Information Retrieval - Kevin P. Murphy — Probabilistic Machine Learning - Selected research papers from conferences such as SIGIR, KDD, WWW, NeurIPS, ICML, and VLDB
[letzte Änderung 16.03.2026]
|