Virtual coach: predict physical activity using a machine learning approach

Research output: Contribution to conferencePaperAcademic

46 Downloads (Pure)

Abstract

One of the main causes of numerous health problems
is a lack of physical activity. To promote a more active lifestyle,
the Hanze University started a health promotion program. Participants were motivated to reach their daily goal of physical activity
by means of an activity tracker in combination with two-weekly
coaching sessions. Employing the data of the experiment, we
investigated the manners in which the predictability of physical
activity of a participant during the day can be improved. The
collected step count data was used to construct personalised
machine learning models, by taking into account the difference
between physical activities during weekdays on the one hand
and weekends on the other hand. The training of algorithms
per participant in combination with the time-slices weekdays,
weekend and the whole week improves the accuracy of the
prediction model. The performance of the models improves
even further when the individualised time-sliced models are
combined. More contextual data, like free time and working
hours, might even extend the accuracy. The use of personalised
prediction models, based on machine learning and time slices,
could become an addition in preventive personalized eHealth
systems and mobile activity monitoring. For instance, this can
constitute as a viable addition to a virtual coaching system to help
the participants to reach their daily goal. As the individualised
models allow for predictions of the progression of the physical
activity during the day, they enable the virtual coaching system
to intervene at the appropriate moment in time.
Original languageEnglish
Number of pages4
Publication statusPublished - 27 Mar 2018
EventeTELEMED2018: 10th International Conference on eHealth, Telemedicine, and Social Medicine 2018 - Rome, Italy
Duration: 25 Mar 201829 Mar 2018
https://www.iaria.org/conferences2018/eTELEMED18.html

Conference

ConferenceeTELEMED2018
CountryItaly
CityRome
Period25/03/1829/03/18
Internet address

Keywords

  • physical activity
  • health
  • coaching
  • ict

Cite this

@conference{d2f2e7f1a3e84df1b7a6d89b551b6944,
title = "Virtual coach: predict physical activity using a machine learning approach",
abstract = "One of the main causes of numerous health problemsis a lack of physical activity. To promote a more active lifestyle,the Hanze University started a health promotion program. Participants were motivated to reach their daily goal of physical activityby means of an activity tracker in combination with two-weeklycoaching sessions. Employing the data of the experiment, weinvestigated the manners in which the predictability of physicalactivity of a participant during the day can be improved. Thecollected step count data was used to construct personalisedmachine learning models, by taking into account the differencebetween physical activities during weekdays on the one handand weekends on the other hand. The training of algorithmsper participant in combination with the time-slices weekdays,weekend and the whole week improves the accuracy of theprediction model. The performance of the models improveseven further when the individualised time-sliced models arecombined. More contextual data, like free time and workinghours, might even extend the accuracy. The use of personalisedprediction models, based on machine learning and time slices,could become an addition in preventive personalized eHealthsystems and mobile activity monitoring. For instance, this canconstitute as a viable addition to a virtual coaching system to helpthe participants to reach their daily goal. As the individualisedmodels allow for predictions of the progression of the physicalactivity during the day, they enable the virtual coaching systemto intervene at the appropriate moment in time.",
keywords = "bewegen (activiteit), coaching, ict, gezondheid, physical activity, health, coaching, ict",
author = "Talko Dijkhuis and Johan Blok and Hugo Velthuijsen",
year = "2018",
month = "3",
day = "27",
language = "English",
note = "null ; Conference date: 25-03-2018 Through 29-03-2018",
url = "https://www.iaria.org/conferences2018/eTELEMED18.html",

}

Virtual coach: predict physical activity using a machine learning approach. / Dijkhuis, Talko; Blok, Johan; Velthuijsen, Hugo.

2018. Paper presented at eTELEMED2018, Rome, Italy.

Research output: Contribution to conferencePaperAcademic

TY - CONF

T1 - Virtual coach: predict physical activity using a machine learning approach

AU - Dijkhuis, Talko

AU - Blok, Johan

AU - Velthuijsen, Hugo

PY - 2018/3/27

Y1 - 2018/3/27

N2 - One of the main causes of numerous health problemsis a lack of physical activity. To promote a more active lifestyle,the Hanze University started a health promotion program. Participants were motivated to reach their daily goal of physical activityby means of an activity tracker in combination with two-weeklycoaching sessions. Employing the data of the experiment, weinvestigated the manners in which the predictability of physicalactivity of a participant during the day can be improved. Thecollected step count data was used to construct personalisedmachine learning models, by taking into account the differencebetween physical activities during weekdays on the one handand weekends on the other hand. The training of algorithmsper participant in combination with the time-slices weekdays,weekend and the whole week improves the accuracy of theprediction model. The performance of the models improveseven further when the individualised time-sliced models arecombined. More contextual data, like free time and workinghours, might even extend the accuracy. The use of personalisedprediction models, based on machine learning and time slices,could become an addition in preventive personalized eHealthsystems and mobile activity monitoring. For instance, this canconstitute as a viable addition to a virtual coaching system to helpthe participants to reach their daily goal. As the individualisedmodels allow for predictions of the progression of the physicalactivity during the day, they enable the virtual coaching systemto intervene at the appropriate moment in time.

AB - One of the main causes of numerous health problemsis a lack of physical activity. To promote a more active lifestyle,the Hanze University started a health promotion program. Participants were motivated to reach their daily goal of physical activityby means of an activity tracker in combination with two-weeklycoaching sessions. Employing the data of the experiment, weinvestigated the manners in which the predictability of physicalactivity of a participant during the day can be improved. Thecollected step count data was used to construct personalisedmachine learning models, by taking into account the differencebetween physical activities during weekdays on the one handand weekends on the other hand. The training of algorithmsper participant in combination with the time-slices weekdays,weekend and the whole week improves the accuracy of theprediction model. The performance of the models improveseven further when the individualised time-sliced models arecombined. More contextual data, like free time and workinghours, might even extend the accuracy. The use of personalisedprediction models, based on machine learning and time slices,could become an addition in preventive personalized eHealthsystems and mobile activity monitoring. For instance, this canconstitute as a viable addition to a virtual coaching system to helpthe participants to reach their daily goal. As the individualisedmodels allow for predictions of the progression of the physicalactivity during the day, they enable the virtual coaching systemto intervene at the appropriate moment in time.

KW - bewegen (activiteit)

KW - coaching

KW - ict

KW - gezondheid

KW - physical activity

KW - health

KW - coaching

KW - ict

M3 - Paper

ER -