The aim of this project is to develop a method for objectively summarising argumentative texts using machine language processing. The goal is to identify the essence of an argument and reproduce it in a concise and unbiased manner. The project is motivated by the idea of argumentation engines that present positions on controversial issues to users in a coherent and understandable way. A prototype search engine, args.me, compares pro and contra arguments on search topics to support self-determined opinion formation. The research focus of the project is on text summarisation and style transfer, with the aim of creating a corpus of manually written argument summaries and developing algorithmic methods for automatically summarising argumentative justifications and neutralising their style. The project will contribute to the goals of RATIO and to natural language processing research in general.
Contact
Project Coordinator and Project Manager