| Autor: |
Nunes, Ingrid Bruno, de Santana, Maíra Araujo, Gomes, Juliana Carneiro, Torcate, Arianne Sarmento, Charron, Nicole, de Brito, Nathália Córdula, Moreno, Giselle Machado Magalhães, de Gusmão, Cristine Martins Gomes, dos Santos, Wellington Pinheiro |
| Zdroj: |
Research on Biomedical Engineering; 20230101, Issue: Preprints p1-11, 11p |
| Abstrakt: |
Purpose: Music accompanies all phases of our lives, and when we reach old age, music becomes a direct symbol of nostalgia. Autobiographical memories are essential to an individual’s sense of identity, continuity, and meaning. But some pathologies, such as dementia, can interrupt the memory storage process. Music can help recall and evoke memories and can be used in alternative treatments for dementia. Methods: This work aims to propose an architecture for a music recommendation system capable of recommending music according to musical genre, with the aim of helping music therapists in therapies aimed at elderly people without dementia or in an initial state of dementia. The public music database Emotify, composed of 400 songs labeled by 1595 participants in 7975 sessions, was used. The songs were organized into two channels of 20 s windows with 5 s overlapping. These windows were represented by 34 time and frequency characteristics. Classifiers based on support vector machines and Random Forests were investigated. Results: The most suitable architecture in this experimental study was the Random Forest with 300 trees, with an accuracy of 72% ± 4%, AUC of 0.89 ± 0.03, sensitivity of 0.68 ± 0.08, and specificity of 0.88 ± 0.03. Conclusions: This exploratory study showed that it is possible to build recommendation systems to support music therapy capable of supporting music therapists based on the automatic classification of songs according to the most appropriate musical genre for the patient, according to previous evaluations carried out by the therapist. |
| Databáze: |
Supplemental Index |
| Externí odkaz: |
|
Full text from SpringerLink