Language Analytics applies a combination of natural language processing, audio analysis and Large Language Models to develop new trustworthy applications. Our focus includes detecting fake news, monitoring disinformation, and developing advanced applications for knowledge extraction.
Our scientific focus lies on new approaches to gain insights from human communication, taking a media holistic approach to the application of audio analysis, natural language processing, and Large Language Models (LLMs). Language can occur in any type of medium. To facilitate complex tasks of DSS stakeholders, it is important to research efficient, accurate, and trustworthy approaches that a) foster understanding/generation of knowledge/insights to solve complex and safety/security-critical tasks. Requirements in this research topic are driven by stakeholder requirements in the public sector, elicited in the context of multiple national projects. We will focus on our leading Austrian expertise in the areas of fake news and disinformation campaign detection and monitoring and audio analysis, such as keyword spotting under noisy conditions. At the same time, we will develop new expertise in the application of local LLMs to applications such as information extraction, summarization, and the semi-automated construction of Knowledge Graphs.
Goals
- Support decision-making processes for stakeholders by creating trustworthy applications.
- Build new expertise in applying Large Language Models (LLMs)
Application Domains
- Disinformation Detection, Hybrid Threats
- Big Data Analytics
