Cluster analysis and artificial neural networks in the characterization and classification of positive mental health profiles in confinement by COVID-19

ABSTRACT

Objective: This work aims to provide a valid procedure to characterize and classify positive mental health profiles of subjects in prolonged confinement by COVID-19. M ethod Methodologically, the factors prosocial attitude, self-control, personal satisfaction, autonomy, problem-solving and self-actualization, and interpersonal relationships were adopted in 285 individuals aged 18 to 75 years;the literature related to positive mental health, the application of Artificial Neural Networks and Cluster Analysis were reviewed, which allowed structuring a systematic analysis procedure to characterize and classify positive mental health profiles. Results: A procedure was achieved that integrates a cluster analysis that identified behavioral patterns and characterized groups of positive mental health and an artificial neural network that classifies positive mental health profiles into two groups distinguishing stages of the well-being of individuals and the normalized importance of the factors analyzed was established. Conclusion: From the results, it can be concluded that the proposed procedure that integrates cluster analysis and artificial neural networks can be used as a characterization and classification tool in studies where there are multiple variables, which complements traditional studies, providing robustness in the analysis. © 2022 Academia Nacional de Medicina. All rights reserved.