Stelar

Newsroom

Correlation Detective:
Efficient multivariate correlation discovery

02.12.2022.

Did you know that the amount of data doubles every 1.5 years? 

As new data science techniques are developed the types of analysis we’re doing on this data are becoming more complex and require more computation time. This is why we need software that can’t only do the analysis, but software that can do it fast. 

In his MSc thesis titled titled “Correlation Detective: Efficient multivariate correlation discovery” for which he received the award of the best MSc thesis at TU/e, Koen Minartz developed an algorithm that can find interesting patterns in big data sets up to 500 times faster than existing methods. This will enable experts to discover complex relationships in a data-driven way. This algorithm will be integrated into STELAR for the discovery of multivariate correlations.

You can access and read the thesis via the following link https://research.tue.nl/en/studentTheses/correlation-detective

STELAR project KICKS- OFF!
From raw data lakes to knowledge lakes​​

The Stelar kick-off meeting was held at the end of September in Athens, Greece hosted by the “Athena” Research and Innovation Centre.

Based on a principle “From raw data lakes to knowledge lakes”, STELAR is on a mission to make agrifood data FAIR and AI-ready with Knowledge Lake Management System. This will be achieved through three pilots covering different stages of the food chains:

–  Risk prevention in food supply lines
–  Early crop growth predictions
–  Timely precision farming interventions

This impact-driven #HorizonEurope project will propel the much-needed change in the way data is used in the agrifood system and we are happy to be a part of this seismic shift.