The Project/ The Challenge
The amount of data is constantly increasing in all areas of the company, from the development department to production. In order to analyze this data effectively, databases must be queried, scripts written for analysis, and illustrations generated.
These tasks require programming skills that medium-sized companies often lack, as they depend on specifically trained employees.
Our Solution
With Text-2-Figure, we use language models (Large Language Models, LLMs) from generative artificial intelligence to simplify data analysis. Our Text-2-Figure solution is a chatbot that acts as a translator and communicates with users in natural language as well as with computers in program code.
Leading large language models (LLMs) enable employees to query databases and evaluate the results.
The Result
The chatbot enables data to be transformed into information that can be used as a basis for decision-making. Our prototype is aimed at scientists without programming skills who would normally need help from an analyst to evaluate their own data.
The prototype enables them to perform database queries and generate images from the results. Users interact with their data in natural language and do not need to know SQL or Python, for example. We refer to this prototype internally as “Text-2-Figure.”
Did you know?
No black box: Full transparency about which query led to which program code and result.
Subsequent refinement of the program code by analysts is possible.
Full local installation possible.
