Next Article in Journal
Design of an Edge-Detection CMOS Image Sensor with Built-in Mask Circuits
Next Article in Special Issue
LARa: Creating a Dataset for Human Activity Recognition in Logistics Using Semantic Attributes
Previous Article in Journal
UAV Positioning Mechanisms in Landing Stations: Classification and Engineering Design Review
Previous Article in Special Issue
Sensory Interactive Table (SIT)—Development of a Measurement Instrument to Support Healthy Eating in a Social Dining Setting
Article

Comparing Person-Specific and Independent Models on Subject-Dependent and Independent Human Activity Recognition Performance

1
Insight Centre for Data Analytics, School of Computer Science and Information Technology, University College Cork, T12 XF62 Cork, Ireland
2
Tyndall National Institute, University College Cork, T12 R5CP Cork, Ireland
3
CONNECT Centre for Future Networks and Communications, Tyndall National Institute, University College Cork, T12 R5CP Cork, Ireland
*
Authors to whom correspondence should be addressed.
This paper is an extended version of our paper published in Scheurer, S.; Tedesco, S.; Brown, K.N.; O’Flynn, B. Subject-dependent and-independent human activity recognition with person-specific and-independent models. In Proceedings of the 6th international Workshop on Sensor-based Activity Recognition and Interaction, Rostock, Germany, 16–17 September 2019; pp. 1–7
Sensors 2020, 20(13), 3647; https://doi.org/10.3390/s20133647
Received: 31 May 2020 / Revised: 23 June 2020 / Accepted: 24 June 2020 / Published: 29 June 2020
(This article belongs to the Special Issue Sensor-Based Activity Recognition and Interaction)
The distinction between subject-dependent and subject-independent performance is ubiquitous in the human activity recognition (HAR) literature. We assess whether HAR models really do achieve better subject-dependent performance than subject-independent performance, whether a model trained with data from many users achieves better subject-independent performance than one trained with data from a single person, and whether one trained with data from a single specific target user performs better for that user than one trained with data from many. To those ends, we compare four popular machine learning algorithms’ subject-dependent and subject-independent performances across eight datasets using three different personalisation–generalisation approaches, which we term person-independent models (PIMs), person-specific models (PSMs), and ensembles of PSMs (EPSMs). We further consider three different ways to construct such an ensemble: unweighted, κ -weighted, and baseline-feature-weighted. Our analysis shows that PSMs outperform PIMs by 43.5% in terms of their subject-dependent performances, whereas PIMs outperform PSMs by 55.9% and κ -weighted EPSMs—the best-performing EPSM type—by 16.4% in terms of the subject-independent performance. View Full-Text
Keywords: human activity recognition; machine learning; ensemble methods; boosting; bagging; inertial sensors human activity recognition; machine learning; ensemble methods; boosting; bagging; inertial sensors
Show Figures

Figure 1

MDPI and ACS Style

Scheurer, S.; Tedesco, S.; O’Flynn, B.; Brown, K.N. Comparing Person-Specific and Independent Models on Subject-Dependent and Independent Human Activity Recognition Performance. Sensors 2020, 20, 3647. https://doi.org/10.3390/s20133647

AMA Style

Scheurer S, Tedesco S, O’Flynn B, Brown KN. Comparing Person-Specific and Independent Models on Subject-Dependent and Independent Human Activity Recognition Performance. Sensors. 2020; 20(13):3647. https://doi.org/10.3390/s20133647

Chicago/Turabian Style

Scheurer, Sebastian, Salvatore Tedesco, Brendan O’Flynn, and Kenneth N. Brown. 2020. "Comparing Person-Specific and Independent Models on Subject-Dependent and Independent Human Activity Recognition Performance" Sensors 20, no. 13: 3647. https://doi.org/10.3390/s20133647

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop