Human Activity Recognition using deep learning techniques
Human activity recognition (HAR) is at the forefront of Pervasive Computing efforts, and deep learning techniques currently empower the most successful endeavors within the field. By using a publicly available dataset an exploratory analysis of feature learning is put forward in this work. The convo...
- Autores:
-
Gómez Meneses, Fabián Andrés
- Tipo de recurso:
- Fecha de publicación:
- 2018
- Institución:
- Universidad Nacional de Colombia
- Repositorio:
- Universidad Nacional de Colombia
- Idioma:
- spa
- OAI Identifier:
- oai:repositorio.unal.edu.co:unal/69588
- Acceso en línea:
- https://repositorio.unal.edu.co/handle/unal/69588
http://bdigital.unal.edu.co/71559/
- Palabra clave:
- 0 Generalidades / Computer science, information and general works
62 Ingeniería y operaciones afines / Engineering
Machine learning
Deep learning
Human behavior
Neural networks
Pervasive computing
Aprendizaje de máquina
Aprendizaje profundo
Comportamiento humano
Redes neuronales
Computación ubicua
- Rights
- openAccess
- License
- Atribución-NoComercial 4.0 Internacional
| Summary: | Human activity recognition (HAR) is at the forefront of Pervasive Computing efforts, and deep learning techniques currently empower the most successful endeavors within the field. By using a publicly available dataset an exploratory analysis of feature learning is put forward in this work. The convolutional neural network deployed here highlights both the advantages and limitations of this class of models, while offering an overview of machine learning-aided human behavior analysis. Furthermore, the exploration includes an experimental comparison with a more traditional SVM model with feature engineering, over the same data. |
|---|