Deep Learning in the Recognition of Activities of Daily Living Using Smartwatch Data
The recognition of human activities (HAR) using wearable device data, such as smartwatches, has gained significant attention in the field of computer science due to its potential to provide insights into individuals’ daily activities. This article aims to conduct a comparative study of deep learning...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-08-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/17/7493 |
| _version_ | 1797581865301835776 |
|---|---|
| author | Ariany F. Cavalcante Victor H. de L. Kunst Thiago de M. Chaves Júlia D. T. de Souza Isabela M. Ribeiro Jonysberg P. Quintino Fabio Q. B. da Silva André L. M. Santos Veronica Teichrieb Alana Elza F. da Gama |
| author_facet | Ariany F. Cavalcante Victor H. de L. Kunst Thiago de M. Chaves Júlia D. T. de Souza Isabela M. Ribeiro Jonysberg P. Quintino Fabio Q. B. da Silva André L. M. Santos Veronica Teichrieb Alana Elza F. da Gama |
| author_sort | Ariany F. Cavalcante |
| collection | DOAJ |
| description | The recognition of human activities (HAR) using wearable device data, such as smartwatches, has gained significant attention in the field of computer science due to its potential to provide insights into individuals’ daily activities. This article aims to conduct a comparative study of deep learning techniques for recognizing activities of daily living (ADL). A mapping of HAR techniques was performed, and three techniques were selected for evaluation, along with a dataset. Experiments were conducted using the selected techniques to assess their performance in ADL recognition, employing standardized evaluation metrics, such as accuracy, precision, recall, and F1-score. Among the evaluated techniques, the DeepConvLSTM architecture, consisting of recurrent convolutional layers and a single LSTM layer, achieved the most promising results. These findings suggest that software applications utilizing this architecture can assist smartwatch users in understanding their movement routines more quickly and accurately. |
| first_indexed | 2024-03-10T23:13:40Z |
| format | Article |
| id | doaj.art-abbe468983ca49b1bb2b040d52398e71 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T23:13:40Z |
| publishDate | 2023-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-abbe468983ca49b1bb2b040d52398e712023-11-19T08:50:37ZengMDPI AGSensors1424-82202023-08-012317749310.3390/s23177493Deep Learning in the Recognition of Activities of Daily Living Using Smartwatch DataAriany F. Cavalcante0Victor H. de L. Kunst1Thiago de M. Chaves2Júlia D. T. de Souza3Isabela M. Ribeiro4Jonysberg P. Quintino5Fabio Q. B. da Silva6André L. M. Santos7Veronica Teichrieb8Alana Elza F. da Gama9Centro de Informática, Universidade Federal de Pernambuco, Recife 50740-560, PE, BrazilCentro de Informática, Universidade Federal de Pernambuco, Recife 50740-560, PE, BrazilCentro de Informática, Universidade Federal de Pernambuco, Recife 50740-560, PE, BrazilCentro de Informática, Universidade Federal de Pernambuco, Recife 50740-560, PE, BrazilCentro de Informática, Universidade Federal de Pernambuco, Recife 50740-560, PE, BrazilProjeto CIn-UFPE Samsung, Centro de Informática, Recife 50740-560, PE, BrazilCentro de Informática, Universidade Federal de Pernambuco, Recife 50740-560, PE, BrazilCentro de Informática, Universidade Federal de Pernambuco, Recife 50740-560, PE, BrazilCentro de Informática, Universidade Federal de Pernambuco, Recife 50740-560, PE, BrazilCentro de Informática, Universidade Federal de Pernambuco, Recife 50740-560, PE, BrazilThe recognition of human activities (HAR) using wearable device data, such as smartwatches, has gained significant attention in the field of computer science due to its potential to provide insights into individuals’ daily activities. This article aims to conduct a comparative study of deep learning techniques for recognizing activities of daily living (ADL). A mapping of HAR techniques was performed, and three techniques were selected for evaluation, along with a dataset. Experiments were conducted using the selected techniques to assess their performance in ADL recognition, employing standardized evaluation metrics, such as accuracy, precision, recall, and F1-score. Among the evaluated techniques, the DeepConvLSTM architecture, consisting of recurrent convolutional layers and a single LSTM layer, achieved the most promising results. These findings suggest that software applications utilizing this architecture can assist smartwatch users in understanding their movement routines more quickly and accurately.https://www.mdpi.com/1424-8220/23/17/7493human activity recognitionneural networkssmartwatchwearable sensor data |
| spellingShingle | Ariany F. Cavalcante Victor H. de L. Kunst Thiago de M. Chaves Júlia D. T. de Souza Isabela M. Ribeiro Jonysberg P. Quintino Fabio Q. B. da Silva André L. M. Santos Veronica Teichrieb Alana Elza F. da Gama Deep Learning in the Recognition of Activities of Daily Living Using Smartwatch Data Sensors human activity recognition neural networks smartwatch wearable sensor data |
| title | Deep Learning in the Recognition of Activities of Daily Living Using Smartwatch Data |
| title_full | Deep Learning in the Recognition of Activities of Daily Living Using Smartwatch Data |
| title_fullStr | Deep Learning in the Recognition of Activities of Daily Living Using Smartwatch Data |
| title_full_unstemmed | Deep Learning in the Recognition of Activities of Daily Living Using Smartwatch Data |
| title_short | Deep Learning in the Recognition of Activities of Daily Living Using Smartwatch Data |
| title_sort | deep learning in the recognition of activities of daily living using smartwatch data |
| topic | human activity recognition neural networks smartwatch wearable sensor data |
| url | https://www.mdpi.com/1424-8220/23/17/7493 |
| work_keys_str_mv | AT arianyfcavalcante deeplearningintherecognitionofactivitiesofdailylivingusingsmartwatchdata AT victorhdelkunst deeplearningintherecognitionofactivitiesofdailylivingusingsmartwatchdata AT thiagodemchaves deeplearningintherecognitionofactivitiesofdailylivingusingsmartwatchdata AT juliadtdesouza deeplearningintherecognitionofactivitiesofdailylivingusingsmartwatchdata AT isabelamribeiro deeplearningintherecognitionofactivitiesofdailylivingusingsmartwatchdata AT jonysbergpquintino deeplearningintherecognitionofactivitiesofdailylivingusingsmartwatchdata AT fabioqbdasilva deeplearningintherecognitionofactivitiesofdailylivingusingsmartwatchdata AT andrelmsantos deeplearningintherecognitionofactivitiesofdailylivingusingsmartwatchdata AT veronicateichrieb deeplearningintherecognitionofactivitiesofdailylivingusingsmartwatchdata AT alanaelzafdagama deeplearningintherecognitionofactivitiesofdailylivingusingsmartwatchdata |