Deep Neural Networks for Human Activity Recognition With Wearable Sensors: Leave-One-Subject-Out Cross-Validation for Model Selection
Human Activity Recognition (HAR) has been attracting significant research attention because of the increasing availability of environmental and wearable sensors for collecting HAR data. In recent years, deep learning approaches have demonstrated a great success due to their ability to model complex...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9144538/ |
| _version_ | 1830081639774945280 |
|---|---|
| author | Davoud Gholamiangonabadi Nikita Kiselov Katarina Grolinger |
| author_facet | Davoud Gholamiangonabadi Nikita Kiselov Katarina Grolinger |
| author_sort | Davoud Gholamiangonabadi |
| collection | DOAJ |
| description | Human Activity Recognition (HAR) has been attracting significant research attention because of the increasing availability of environmental and wearable sensors for collecting HAR data. In recent years, deep learning approaches have demonstrated a great success due to their ability to model complex systems. However, these models are often evaluated on the same subjects as those used to train the model; thus, the provided accuracy estimates do not pertain to new subjects. Occasionally, one or a few subjects are selected for the evaluation, but such estimates highly depend on the subjects selected for the evaluation. Consequently, this paper examines how well different machine learning architectures make generalizations based on a new subject(s) by using Leave-One-Subject-Out Cross-Validation (LOSOCV). Changing the subject used for the evaluation in each fold of the cross-validation, LOSOCV provides subject-independent estimate of the performance for new subjects. Six feed forward and convolutional neural network (CNN) architectures as well as four pre-processing scenarios have been considered. Results show that CNN architecture with two convolutions and one-dimensional filter accompanied by a sliding window and vector magnitude, generalizes better than other architectures. For the same CNN, the accuracy improves from 85.1% when evaluated with LOSOCV to 99.85% when evaluated with the traditional 10-fold cross-validation, demonstrating the importance of using LOSOCV for the evaluation. |
| first_indexed | 2024-12-14T15:44:37Z |
| format | Article |
| id | doaj.art-9bea9b3dbf2c4991ab92d75c8c999e3c |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-14T15:44:37Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-9bea9b3dbf2c4991ab92d75c8c999e3c2022-12-21T22:55:32ZengIEEEIEEE Access2169-35362020-01-01813398213399410.1109/ACCESS.2020.30107159144538Deep Neural Networks for Human Activity Recognition With Wearable Sensors: Leave-One-Subject-Out Cross-Validation for Model SelectionDavoud Gholamiangonabadi0https://orcid.org/0000-0002-5476-0959Nikita Kiselov1https://orcid.org/0000-0003-0297-7811Katarina Grolinger2https://orcid.org/0000-0003-0062-8212Department of Electrical and Computer Engineering, Western University, London, CanadaDepartment of Electrical and Computer Engineering, Western University, London, CanadaDepartment of Electrical and Computer Engineering, Western University, London, CanadaHuman Activity Recognition (HAR) has been attracting significant research attention because of the increasing availability of environmental and wearable sensors for collecting HAR data. In recent years, deep learning approaches have demonstrated a great success due to their ability to model complex systems. However, these models are often evaluated on the same subjects as those used to train the model; thus, the provided accuracy estimates do not pertain to new subjects. Occasionally, one or a few subjects are selected for the evaluation, but such estimates highly depend on the subjects selected for the evaluation. Consequently, this paper examines how well different machine learning architectures make generalizations based on a new subject(s) by using Leave-One-Subject-Out Cross-Validation (LOSOCV). Changing the subject used for the evaluation in each fold of the cross-validation, LOSOCV provides subject-independent estimate of the performance for new subjects. Six feed forward and convolutional neural network (CNN) architectures as well as four pre-processing scenarios have been considered. Results show that CNN architecture with two convolutions and one-dimensional filter accompanied by a sliding window and vector magnitude, generalizes better than other architectures. For the same CNN, the accuracy improves from 85.1% when evaluated with LOSOCV to 99.85% when evaluated with the traditional 10-fold cross-validation, demonstrating the importance of using LOSOCV for the evaluation.https://ieeexplore.ieee.org/document/9144538/Deep neural networkshuman activity recognitionmodel selectionconvolutional neural networksfeed forward neural networksmodel evaluation |
| spellingShingle | Davoud Gholamiangonabadi Nikita Kiselov Katarina Grolinger Deep Neural Networks for Human Activity Recognition With Wearable Sensors: Leave-One-Subject-Out Cross-Validation for Model Selection IEEE Access Deep neural networks human activity recognition model selection convolutional neural networks feed forward neural networks model evaluation |
| title | Deep Neural Networks for Human Activity Recognition With Wearable Sensors: Leave-One-Subject-Out Cross-Validation for Model Selection |
| title_full | Deep Neural Networks for Human Activity Recognition With Wearable Sensors: Leave-One-Subject-Out Cross-Validation for Model Selection |
| title_fullStr | Deep Neural Networks for Human Activity Recognition With Wearable Sensors: Leave-One-Subject-Out Cross-Validation for Model Selection |
| title_full_unstemmed | Deep Neural Networks for Human Activity Recognition With Wearable Sensors: Leave-One-Subject-Out Cross-Validation for Model Selection |
| title_short | Deep Neural Networks for Human Activity Recognition With Wearable Sensors: Leave-One-Subject-Out Cross-Validation for Model Selection |
| title_sort | deep neural networks for human activity recognition with wearable sensors leave one subject out cross validation for model selection |
| topic | Deep neural networks human activity recognition model selection convolutional neural networks feed forward neural networks model evaluation |
| url | https://ieeexplore.ieee.org/document/9144538/ |
| work_keys_str_mv | AT davoudgholamiangonabadi deepneuralnetworksforhumanactivityrecognitionwithwearablesensorsleaveonesubjectoutcrossvalidationformodelselection AT nikitakiselov deepneuralnetworksforhumanactivityrecognitionwithwearablesensorsleaveonesubjectoutcrossvalidationformodelselection AT katarinagrolinger deepneuralnetworksforhumanactivityrecognitionwithwearablesensorsleaveonesubjectoutcrossvalidationformodelselection |