A Hybrid Machine Learning Framework for Enhancing the Prediction Power in Large Scale Population Studies: The ATHLOS Project [article]

Petros Barmpas, Sotiris Tasoulis, Aristidis G. Vrahatis, Matthew Prina, José Luis Ayuso-Mateos, Jerome Bickenbach, Ivet Bayes, Martin Bobak, Francisco Félix Caballero, Somnath Chatterji, Laia Egea-Cortés, Esther García-Esquinas (+13 others)
2021 medRxiv   pre-print
AbstractThe ATHLOS cohort is composed of several harmonized datasets of international cohorts related to health and aging. The healthy aging scale has been constructed based on a selection of particular variables from 16 individual studies. In this paper, we consider a selection of additional variables found in ATHLOS and investigate their utilization for predicting the healthy aging. For this purpose motivated by the dataset's volume and diversity we focus our attention upon the clustering for
more » ... prediction scheme, where unsupervised learning is utilized to enhance prediction power, showing the predictive utility of exploiting structure in the data by clustering. We show that imposed computation bottlenecks can be surpassed when using appropriate hierarchical clustering within a clustering for ensemble classification scheme while retaining prediction benefits. We propose a complete methodology which is evaluated against baseline methods and the original concept. The results are very encouraging suggesting further developments in this direction along with applications in tasks with similar characteristics. A strait-forward open source implementation is provided for the R project.
doi:10.1101/2021.01.23.21250355 fatcat:wybczqevgvg3bkvh4pwv36znxe