Comparing model performances applied to fall detection

1. Barri Khojasteh, Samad
2. Villar Flecha, José Ramón
3. Cal Marín, Enrique Antonio de la
4. González Suárez, Víctor Manuel
5. Chira, Camelia

Exportar

Exportar

×

MLA

modern-language-association

APA

apa

Chicago

chicago-fullnote-bibliography

Harvard

harvard-cite-them-right

Vancouver

vancouver-author-date

---

RIS BibTex

Exportar

×

El documento no se puede exportar porque pertenece a otro portal.

Texto completo

lock_openRUO externo lock_openRUO externo

GOOGLE SCHOLAR lock_openRUO editor

Resumen

This study focuses on the comparison of techniques for modelling and classifying data gathered from wearable sensors, in order to detect fall events of elderly people. Although the vast majority of studies concerning fall detection place the sensory on the waist, in this research the wearable device must be placed on the wrist because it’s usability. A first pre-processing stage is carried out as stated in [1], [2]; this stage detects the most relevant points to label. This study analyses the suitability of different models in solving this classification problem: a feed-forward Neural Network and a decision tree based on C5.0. A discussion about the results and the deployment issues is performed according to whether the models are to be exploited in edge/cloud computing or in the wearable device.