학술논문
A divisive hierarchical clustering methodology for enhancing the ensemble prediction power in large scale population studies: the ATHLOS project
Document Type
Report
Author
Barmpas, Petros; Tasoulis, Sotiris; Vrahatis, Aristidis G.; Georgakopoulos, Spiros V.; Anagnostou, Panagiotis; Prina, Matthew; Ayuso-Mateos, José Luis; Bickenbach, Jerome; Bayes, Ivet; Bobak, Martin; Caballero, Francisco Félix; Chatterji, Somnath; Egea-Cortés, Laia; García-Esquinas, Esther; Leonardi, Matilde; Koskinen, Seppo; Koupil, Ilona; Paja̧k, Andrzej; Prince, Martin; Sanderson, Warren; Scherbov, Sergei; Tamosiunas, Abdonas; Galas, Aleksander; Haro, Josep Maria; Sanchez-Niubo, Albert; Plagianakos, Vassilis P.; Panagiotakos, Demosthenes
Source
Health Information Science and Systems. April 18, 2022, Vol. 10 Issue 1
Subject
Language
English
ISSN
2047-2501
Abstract
Author(s): Petros Barmpas[sup.1], Sotiris Tasoulis[sup.1], Aristidis G. Vrahatis[sup.1], Spiros V. Georgakopoulos[sup.27], Panagiotis Anagnostou[sup.1], Matthew Prina[sup.2,3], José Luis Ayuso-Mateos[sup.4,5,6], Jerome Bickenbach[sup.7,8], Ivet Bayes[sup.4,13], Martin Bobak[sup.9], Francisco Félix Caballero[sup.10,11], Somnath Chatterji[sup.12], Laia [...]
The ATHLOS cohort is composed of several harmonized datasets of international groups related to health and aging. As a result, the Healthy Aging index has been constructed based on a selection of variables from 16 individual studies. In this paper, we consider additional variables found in ATHLOS and investigate their utilization for predicting the Healthy Aging index. For this purpose, motivated by the volume and diversity of the dataset, we focus our attention upon data clustering, where unsupervised learning is utilized to enhance prediction power. Thus we show the predictive utility of exploiting hidden data structures. In addition, we demonstrate 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 that 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 straightforward open source implementation for the R project is also provided (https://github.com/Petros-Barmpas/HCEP). Keywords: Clustering, Prediction enhancement, ATHLOS cohort, Ensemble methods
The ATHLOS cohort is composed of several harmonized datasets of international groups related to health and aging. As a result, the Healthy Aging index has been constructed based on a selection of variables from 16 individual studies. In this paper, we consider additional variables found in ATHLOS and investigate their utilization for predicting the Healthy Aging index. For this purpose, motivated by the volume and diversity of the dataset, we focus our attention upon data clustering, where unsupervised learning is utilized to enhance prediction power. Thus we show the predictive utility of exploiting hidden data structures. In addition, we demonstrate 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 that 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 straightforward open source implementation for the R project is also provided (https://github.com/Petros-Barmpas/HCEP). Keywords: Clustering, Prediction enhancement, ATHLOS cohort, Ensemble methods