Using Accelerometer and GPS Data for Real-Life Physical Activity Type Detection
This paper aims to examine the role of global positioning system (GPS) sensor data in real-life physical activity (PA) type detection. Thirty-three young participants wore devices including GPS and accelerometer sensors on five body positions and performed daily PAs in two protocols, namely semi-str...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-01-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/20/3/588 |
_version_ | 1798038990434074624 |
---|---|
author | Hoda Allahbakhshi Lindsey Conrow Babak Naimi Robert Weibel |
author_facet | Hoda Allahbakhshi Lindsey Conrow Babak Naimi Robert Weibel |
author_sort | Hoda Allahbakhshi |
collection | DOAJ |
description | This paper aims to examine the role of global positioning system (GPS) sensor data in real-life physical activity (PA) type detection. Thirty-three young participants wore devices including GPS and accelerometer sensors on five body positions and performed daily PAs in two protocols, namely semi-structured and real-life. One general random forest (RF) model integrating data from all sensors and five individual RF models using data from each sensor position were trained using semi-structured (Scenario 1) and combined (semi-structured + real-life) data (Scenario 2). The results showed that in general, adding GPS features (speed and elevation difference) to accelerometer data improves classification performance particularly for detecting non-level and level walking. Assessing the transferability of the models on real-life data showed that models from Scenario 2 are strongly transferable, particularly when adding GPS data to the training data. Comparing individual models indicated that knee-models provide comparable classification performance (above 80%) to general models in both scenarios. In conclusion, adding GPS data improves real-life PA type classification performance if combined data are used for training the model. Moreover, the knee-model provides the minimal device configuration with reliable accuracy for detecting real-life PA types. |
first_indexed | 2024-04-11T21:47:46Z |
format | Article |
id | doaj.art-56317111b88e4a329f31349b80ad8658 |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-04-11T21:47:46Z |
publishDate | 2020-01-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-56317111b88e4a329f31349b80ad86582022-12-22T04:01:21ZengMDPI AGSensors1424-82202020-01-0120358810.3390/s20030588s20030588Using Accelerometer and GPS Data for Real-Life Physical Activity Type DetectionHoda Allahbakhshi0Lindsey Conrow1Babak Naimi2Robert Weibel3Department of Geography, Geographic Information Systems Unit, University of Zurich (UZH), Winterthurerstrasse 190, 8057 Zurich, SwitzerlandDepartment of Geography, Geographic Information Systems Unit, University of Zurich (UZH), Winterthurerstrasse 190, 8057 Zurich, SwitzerlandDepartment of Geosciences and Geography, University of Helsinki, P.O. Box 64, 00014 Helsinki, FinlandDepartment of Geography, Geographic Information Systems Unit, University of Zurich (UZH), Winterthurerstrasse 190, 8057 Zurich, SwitzerlandThis paper aims to examine the role of global positioning system (GPS) sensor data in real-life physical activity (PA) type detection. Thirty-three young participants wore devices including GPS and accelerometer sensors on five body positions and performed daily PAs in two protocols, namely semi-structured and real-life. One general random forest (RF) model integrating data from all sensors and five individual RF models using data from each sensor position were trained using semi-structured (Scenario 1) and combined (semi-structured + real-life) data (Scenario 2). The results showed that in general, adding GPS features (speed and elevation difference) to accelerometer data improves classification performance particularly for detecting non-level and level walking. Assessing the transferability of the models on real-life data showed that models from Scenario 2 are strongly transferable, particularly when adding GPS data to the training data. Comparing individual models indicated that knee-models provide comparable classification performance (above 80%) to general models in both scenarios. In conclusion, adding GPS data improves real-life PA type classification performance if combined data are used for training the model. Moreover, the knee-model provides the minimal device configuration with reliable accuracy for detecting real-life PA types.https://www.mdpi.com/1424-8220/20/3/588physical activity typereal-lifegpsgis |
spellingShingle | Hoda Allahbakhshi Lindsey Conrow Babak Naimi Robert Weibel Using Accelerometer and GPS Data for Real-Life Physical Activity Type Detection Sensors physical activity type real-life gps gis |
title | Using Accelerometer and GPS Data for Real-Life Physical Activity Type Detection |
title_full | Using Accelerometer and GPS Data for Real-Life Physical Activity Type Detection |
title_fullStr | Using Accelerometer and GPS Data for Real-Life Physical Activity Type Detection |
title_full_unstemmed | Using Accelerometer and GPS Data for Real-Life Physical Activity Type Detection |
title_short | Using Accelerometer and GPS Data for Real-Life Physical Activity Type Detection |
title_sort | using accelerometer and gps data for real life physical activity type detection |
topic | physical activity type real-life gps gis |
url | https://www.mdpi.com/1424-8220/20/3/588 |
work_keys_str_mv | AT hodaallahbakhshi usingaccelerometerandgpsdataforreallifephysicalactivitytypedetection AT lindseyconrow usingaccelerometerandgpsdataforreallifephysicalactivitytypedetection AT babaknaimi usingaccelerometerandgpsdataforreallifephysicalactivitytypedetection AT robertweibel usingaccelerometerandgpsdataforreallifephysicalactivitytypedetection |