Training data selection with user's physical characteristics data for acceleration-based activity modeling

Takuya Maekawa, Shinji Watanabe

Research output: Contribution to journalArticle

Abstract

This paper proposes an activity recognition method that models an end user's activities without using any labeled/unlabeled acceleration sensor data obtained from the user. Our method employs information about the end user's physical characteristics such as height and gender to find and select appropriate training data obtained from other users in advance. Then, we model the end user's activities by using the selected labeled sensor data. Therefore, our method does not require the end user to collect and label her training sensor data. In this paper, we propose and test two methods for finding appropriate training data by using information about the end user's physical characteristics. Moreover, our recognition method improves the recognition performance without the need for any effort by the end user because the method automatically adapts the activity models to the end user when it recognizes her unlabeled sensor data. We confirmed the effectiveness of our method by using 100 h of sensor data obtained from 40 participants.

Original languageEnglish
Pages (from-to)451-463
Number of pages13
JournalPersonal and Ubiquitous Computing
Volume17
Issue number3
DOIs
Publication statusPublished - 2013 Mar
Externally publishedYes

Fingerprint

Sensors
Labels
End users
Modeling
Sensor

Keywords

  • Acceleration sensor
  • Activity recognition
  • Physical characteristics
  • Wearable sensors

ASJC Scopus subject areas

  • Hardware and Architecture
  • Computer Science Applications
  • Management Science and Operations Research

Cite this

Training data selection with user's physical characteristics data for acceleration-based activity modeling. / Maekawa, Takuya; Watanabe, Shinji.

In: Personal and Ubiquitous Computing, Vol. 17, No. 3, 03.2013, p. 451-463.

Research output: Contribution to journalArticle

@article{5a209baef45e4948b82f16681e8bdcee,
title = "Training data selection with user's physical characteristics data for acceleration-based activity modeling",
abstract = "This paper proposes an activity recognition method that models an end user's activities without using any labeled/unlabeled acceleration sensor data obtained from the user. Our method employs information about the end user's physical characteristics such as height and gender to find and select appropriate training data obtained from other users in advance. Then, we model the end user's activities by using the selected labeled sensor data. Therefore, our method does not require the end user to collect and label her training sensor data. In this paper, we propose and test two methods for finding appropriate training data by using information about the end user's physical characteristics. Moreover, our recognition method improves the recognition performance without the need for any effort by the end user because the method automatically adapts the activity models to the end user when it recognizes her unlabeled sensor data. We confirmed the effectiveness of our method by using 100 h of sensor data obtained from 40 participants.",
keywords = "Acceleration sensor, Activity recognition, Physical characteristics, Wearable sensors",
author = "Takuya Maekawa and Shinji Watanabe",
year = "2013",
month = "3",
doi = "10.1007/s00779-011-0491-0",
language = "English",
volume = "17",
pages = "451--463",
journal = "Personal and Ubiquitous Computing",
issn = "1617-4909",
publisher = "Springer London",
number = "3",

}

TY - JOUR

T1 - Training data selection with user's physical characteristics data for acceleration-based activity modeling

AU - Maekawa, Takuya

AU - Watanabe, Shinji

PY - 2013/3

Y1 - 2013/3

N2 - This paper proposes an activity recognition method that models an end user's activities without using any labeled/unlabeled acceleration sensor data obtained from the user. Our method employs information about the end user's physical characteristics such as height and gender to find and select appropriate training data obtained from other users in advance. Then, we model the end user's activities by using the selected labeled sensor data. Therefore, our method does not require the end user to collect and label her training sensor data. In this paper, we propose and test two methods for finding appropriate training data by using information about the end user's physical characteristics. Moreover, our recognition method improves the recognition performance without the need for any effort by the end user because the method automatically adapts the activity models to the end user when it recognizes her unlabeled sensor data. We confirmed the effectiveness of our method by using 100 h of sensor data obtained from 40 participants.

AB - This paper proposes an activity recognition method that models an end user's activities without using any labeled/unlabeled acceleration sensor data obtained from the user. Our method employs information about the end user's physical characteristics such as height and gender to find and select appropriate training data obtained from other users in advance. Then, we model the end user's activities by using the selected labeled sensor data. Therefore, our method does not require the end user to collect and label her training sensor data. In this paper, we propose and test two methods for finding appropriate training data by using information about the end user's physical characteristics. Moreover, our recognition method improves the recognition performance without the need for any effort by the end user because the method automatically adapts the activity models to the end user when it recognizes her unlabeled sensor data. We confirmed the effectiveness of our method by using 100 h of sensor data obtained from 40 participants.

KW - Acceleration sensor

KW - Activity recognition

KW - Physical characteristics

KW - Wearable sensors

UR - http://www.scopus.com/inward/record.url?scp=84879690947&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84879690947&partnerID=8YFLogxK

U2 - 10.1007/s00779-011-0491-0

DO - 10.1007/s00779-011-0491-0

M3 - Article

VL - 17

SP - 451

EP - 463

JO - Personal and Ubiquitous Computing

JF - Personal and Ubiquitous Computing

SN - 1617-4909

IS - 3

ER -