GENIAL4KMU: Generative information retrieval augmented solutions for unstructured data in SMEs
The project Generative Information Retrieval Augmented Solutions for Unstructured Data in SMEs (GENIAL4KMU) researches the possibilities of making unstructured and structured data more easily and quickly accessible.
The rapid development of artificial intelligence and the publication of ChatGPT and other Large Language Models (LLMs) have made it possible to make company data more accessible. LLMs can solve complex problems and process information from documents.
In addition, research into so-called retrieval-augmented solutions has led to information from other data sources such as documents, photos, or data from databases being linked to LLMs. This connection will be further researched in this project and a practical system for the German market for small and medium-sized enterprises will be designed to make this technology usable. The focus is on the development of German-language systems and practicable role-based access control systems for easy use.
Thematically, the project is divided into two phases, so that initially a system for structured data sources such as databases or tables will be designed. The second phase of the project will focus more on unstructured data stocks so that documents or reports can also be made accessible.
The following project partners are involved in the project:
- Universität Hamburg HCDS
- dida Datenschmiede GmbH
Contact
Funding
BMBF Logo
The research and development project is funded by the German Federal Ministry of Education and Research (BMBF) and supervised by the Deutsches Zentrum für Luft- und Raumfahrt e. V. (DLR).
Funding code of the University of Hamburg: 01IS24044B