Open Lab Night_Mining Arguments from Podcasts
Project information
submitted by
Johannes Kiesel
Co-Authors
Mariia Ufimtseva, Pranay Karani
Mentors
Nailia Mirzakhmedova, Matti Wiegmann, Johannes Kiesel, Benno Stein
Faculty:
Media
Degree programme:
Computer Science and Media (english) (Master of Science (M.Sc.)),
Human-Computer Interaction (Master of Science (M.Sc.)),
Informatik (Bachelor of Science (B.Sc.))
Type of project presentation
Research project
Semester
Summer semester 2023
- Schwanseestraße 143 - S143
(R2.33)
Project description
When faced with a controversial issue for the first time (e.g., should we force people to replace their gas heaters?), it is difficult to get an overview of the main arguments on both sides. At our department, we are developing argumentation technology to find the most important pro and con arguments on each controversial topic. To gather these arguments, we have so far largely relied on well-structured debates from specialized debating websites. However, these websites are often not up to date and lack many real-life arguments. Podcasts would be a great resource to gather diverse and timely topics, but they are often messy and hard to process, so no one has tried to extract arguments from podcasts yet. With this project, we want to open up podcasts as a source for arguments.
In our exhibition, you will learn more about the challenges and opportunities of extracting arguments from podcasts and how we are using state-of-the-art natural language processing methods to overcome these challenges.