iBIGWorld on ICICT’2021

The results of the project iBIGWorld have been presented at the 6th International Congress on Information and Communication Technology (ICICT’2021), February 25-26 London

The work was co-funded by the European Union’s Erasmus + Programme for Education under KA2 grant (project no. 2020-1-PL01-KA203-082197 “Innovations for Big Data in a Real World”)​
The conference was held through digital platform ZOOM.
The work considers ML problems in medical application and presents a minimax approach for developing ML models that would be resistant to aleatoric and epistemic uncertainties.
The main methods applied are based on linear regression, SVM, random forest for ML, PCA for dimension reduction, cross-validation as a resampling strategy.
The approach which is offered is presented with the help of the flowchart which includes basic steps of ML model development under uncertainties, including import and primary processing the clinical data, the statement of task, resampling strategy including the dimension reduction, the choice of methods (learners), tuning their parameters and models comparison on the basis of minimax criterion.
The work is concerned with the iBIGWorld project since it will be used for the development of tutorials for BidData Analytics.