Name of the participant: Tilman Beck
Description of the IT research project: With the rapid development in science over the last decades, the number of scientific publications at conferences, in journals and on other publication platforms is also increasing. This amount of information poses particular challenges for young scientists, as it is a very time-consuming process to identify relevant information. This is especially true for interdisciplinary research, as it combines the research work of several scientific disciplines. In this research project, we focus on the development of Argument Mining (AM) techniques from an information-seeking perspective to identify relevant arguments in a large amount of scientific literature.
The overall objective of this project is to facilitate the search for relevant information for researchers. Similar to a search engine, users should be able to identify relevant arguments in vast amounts of interdisciplinary, scientific literature and thus gain an overview of the discourse in related works. The aim is to use methods of natural language processing:
- to aggregate topic-relevant arguments in a large amount of scientific literature
- to identify redundant contents and group them according to thematic aspects
- to integrate the functionalities into a prototype
Within the framework of this project, we are continuously evaluating the developed models in practice. The prototype should enable researchers to search for relevant arguments in the extensive scientific literature. With the help of this prototype we will collect feedback on the quality of the search results and adapt our models accordingly.
Software Campus partners: TU Darmstadt, Holtzbrinck Publishing Group
Implementation period: 01.01.2020 – 01.01.2022