Cognitive Architecture for Process industries
Presenter: Peter Bednar
Associate professor at the Technical University of Košice, Slovakia
Abstract: This paper introduces a Cross-Sectorial Big Data Processing platform which provides tools for the semantic modelling of the data analytical processes and for the automatic generation of data analysis scripts for solving the described problems.
The main contribution of this paper is the cognitive component for the automatic extraction of the task definition from the narrative description of the problem based on the Large Language Models (LLMs). We have evaluated the proposed method on five problems from the different domains and found that the automatic extraction of the task definition can have promising results that can be applied to full-automatic data analytics.
eSAAM 2023: https://events.eclipse.org/2023/esaam2023/
Sponsored by:
- https://he-codeco.eu/
- https://eucloudedgeiot.eu/
- https://meta-os.eu/
- https://nephele-project.eu/
- https://eclipse.org/