Rapid Artificial Intelligence based Detection of Aggressive or Radical content on the Web (RAIDAR)

The RAIDAR project investigates research methods and approaches for the quantitative collection and evaluation of hate content and radicalization on the web, both which pose a threat to democracy. Further goals are the development of a data science platform for the semi-automated and versatile analysis of large data sets from different sources, as well as the research of approaches and methods for the automated classification of content according to legal paragraphs, which are to be assigned to hate on the web and radicalization from a criminal law perspective. Due to the legal, social and cultural complexity of these tasks and objectives, RAIDAR research includes a comprehensive ethical and legal evaluation.

The innovations of RAIDAR include the development and definition of metrics, measures and methods for quantitative and qualitative evaluation of online hate and radicalization, as well as the application of LegalAI to the application area of Hate Speech. Project results will include partially automated assistance systems based on artificial intelligence in the legal field, a concrete technology assessment of ethical limits and legal frameworks in the context of artificial intelligence for automated data collection, and the application of the RAIDAR platform in a quantitative study in the area of "Hate Speech" and "radicalization" on temporally and contextually relevant content.

Facts:

• Projektbeginn: October 2021
• Projektdauer: 2 years
• Budget: ca. 400k EUR
• Förderung: FFG

Webpage: TBA