Jump to navigation Jump to content Jump to help Jump to language Jump to logout

482372-FS2023-0-Text as Data and Automated Content Analysis





Root number 482372
Semester FS2023
Type of course Block Course
Allocation to subject Communication and Media Sciences
Type of exam Seminar paper
Title Text as Data and Automated Content Analysis
Description Lectured by Ernesto de León

How can computational tools be used to extract meaning and measurements from large volumes of text? This blockseminar focuses on providing students with the tools to conduct text-as-data analysis, allowing them to analyze large corpora of documents quickly and reliably through the R software tool. While the practical examples will focus on classical communication science topics (such as classifying news articles by topic and extracting sentiment from tweets), such tools can be applied to a wide range of settings, including commercial online reviews and legal documents.

Practically, students will first be introduced to the basic functioning of R, followed by simpler automated content analysis tools such as dictionaries and sentiment analysis, to then explore the more complex applied supervised and unsupervised machine learning methods. While no previous experience with R is needed, any experience with coding, in general, is helpful.


Inscription:
from January 15th 2023 (08:00 p.m.) onwards via ILIAS

Form of Implementation: Present
ILIAS-Link (Learning resource for course) No registration/deregistration in CTS (Admission in ILIAS possible). ILIAS
Link to another web site
Lecturers Prof. Dr. Silke Adam, Institute of Communication and Media Studies (icmb) ✉
ECTS 6
Recognition as optional course possible No
Grading 1 to 6
 
Dates Monday 20/2/2023 09:15-17:00
Tuesday 21/2/2023 09:15-17:00
Wednesday 22/2/2023 09:15-17:00
Thursday 23/2/2023 09:15-17:00
Friday 24/2/2023 09:15-17:00
Wednesday 15/3/2023 23:50-23:55
 
Rooms Seminarraum 324, Parkterrasse 14
 
Students please consult the detailed view for complete information on dates, rooms and planned podcasts.