Lellis Moreira MÂ, Simões Gomes CF, Dos Santos M, da Silva Júnior AC, de Araújo Costa IP. Sensitivity Analysis by the PROMETHEE-GAIA method: Algorithms evaluation for COVID-19 prediction.
PROCEDIA COMPUTER SCIENCE 2022;
199:431-438. [PMID:
35136460 PMCID:
PMC8812089 DOI:
10.1016/j.procs.2022.01.052]
[Citation(s) in RCA: 4] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Subscribe] [Scholar Register] [Indexed: 11/19/2022]
Abstract
With the expansion of coronavirus in the World, the search for technology solutions based on the analysis and prospecting of diseases has become constant. The paper addresses a machine learning algorithms analysis used to predict and identify infected patients. For analysis, we use a multicriteria approach using the PROMETHEE-GAIA method, providing the structuring of alternatives respective to a set of criteria, thus enabling the obtaining of their importance degree under the perspective of multiple criteria. The study approaches a sensitivity analysis, evaluating the alternatives using the PROMETHEE I and II methods, along with the GAIA plan, both implemented by the Visual PROMETHEE computational tool, exploring numerical and graphical resources. The analysis model proves to be effective, guaranteeing the ranking of alternatives by inter criterion evaluation and local results with intra criterion evaluation, providing a transparent analysis concerning the selection of prediction algorithms to combat the COVID-19 pandemic.
Collapse