infDLfNLP-01a: Deep Learning for Natural Language Processing (infDLfNLP-01a) (080085)- Dozent/in
- Prof. Dr. Ralf Krestel
- Angaben
- Vorlesung, 2 SWS, ECTS-Studium, ECTS-Credits: 6
Unterrichtssprache Englisch
Zeit und Ort: Do 10:15 - 11:45, LMS8 - R.EG.016 (40)
vom 14.4.2024 bis zum 14.7.2024
1. Prüfungstermin (Klausur am Ende der Vorlesungszeit eines Semesters): 24.7.2024
- Zusätzliche Informationen
- Erwartete Teilnehmerzahl: 25
- Zugeordnete Lehrveranstaltungen
- UE: Exercise: Deep Learning for Natural Language Processing (080087)
-
Dozent/in: Aftab Anjum
Zeit und Ort: Mo 16:15 - 17:45, LMS8 - R.EG.016 (40)
1. Prüfungstermin (Klausur am Ende der Vorlesungszeit eines Semesters): 24.7.2024, 10:00 - 12:00 Uhr, Raum CAP2 - Hörsaal A
infSemDL: Master Seminar - Digital Libraries (infSemDL) (080136)- Dozentinnen/Dozenten
- Prof. Dr. Ralf Krestel, Aftab Anjum
- Angaben
- Seminar, 2 SWS, ECTS-Studium, ECTS-Credits: 5
Unterrichtssprache Englisch
Zeit und Ort: Mo 10:15 - 11:45, LMS8 - R.EG.010 (26)
vom 14.4.2024 bis zum 14.7.2024
- Voraussetzungen / Organisatorisches
- a few lectures
presentations by peers
- Inhalt
- Large document collections, such as the web, are interesting for research from two points of view: first, as a field of study, i.e., to analyze, explore, classify, and recommend or short as an application domain; and second, as a source to train models or discover knowledge. In this seminar, we will focus on one of various aspects of working with large document collections: from large language models to generate text or answer questions, to learning topic models and constructing knowledge graphs.
Lernziele:
Find and assess good research publications
Familiarize with recent research results
Design and execute experiments
Write a research paper
Present your insights
Lehrinhalte:
The topic will be related to digital libraries and large document collections, alternating between language modeling, topic modeling, and knowledge graphs.
- Empfohlene Literatur
- A seed set of literature will be given at the beginning of the seminar.
- Zusätzliche Informationen
- Erwartete Teilnehmerzahl: 10
|
|
