Mobile activity recognition for a whole day: Recognizing real nursing activities with big dataset

Sozo Inoue, Naonori Ueda, Yasunobu Nohara, Naoki Nakashima

研究成果: 著書/レポートタイプへの貢献会議での発言

27 引用 (Scopus)

抄録

In this paper, we provide a real nursing data set for mobile activity recognition that can be used for supervised machine learning, and big data combined the patient medical records and sensors attempted for 2 years, and also propose a method for recognizing activities for a whole day utilizing prior knowledge about the activity segments in a day. Furthermore, we demonstrate data mining by applying our method to the bigger data with additional hospital data. In the proposed method, we 1) convert a set of segment timestamps into a prior probability of the activity segment by exploiting the concept of importance sampling, 2) obtain the likelihood of traditional recognition methods for each local time window within the segment range, and, 3) apply Bayesian estimation by marginalizing the conditional probability of estimating the activities for the segment samples. By evaluating with the dataset, the proposed method outperformed the traditional method without using the prior knowledge by 25.81% at maximum by balanced classification rate. Moreover, the proposed method significantly reduces duration errors of activity segments from 324.2 seconds of the traditional method to 74.6 seconds at maximum. We also demonstrate the data mining by applying our method to bigger data in a hospital.

元の言語英語
ホスト出版物のタイトルUbiComp 2015 - Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing
出版者Association for Computing Machinery, Inc
ページ1269-1280
ページ数12
ISBN(電子版)9781450335744
DOI
出版物ステータス出版済み - 9 7 2015
イベント3rd ACM International Joint Conference on Pervasive and Ubiquitous Computing, UbiComp 2015 - Osaka, 日本
継続期間: 9 7 20159 11 2015

出版物シリーズ

名前UbiComp 2015 - Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing

その他

その他3rd ACM International Joint Conference on Pervasive and Ubiquitous Computing, UbiComp 2015
日本
Osaka
期間9/7/159/11/15

Fingerprint

Nursing
Data mining
Importance sampling
Learning systems
Sensors
Big data

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Hardware and Architecture
  • Software

これを引用

Inoue, S., Ueda, N., Nohara, Y., & Nakashima, N. (2015). Mobile activity recognition for a whole day: Recognizing real nursing activities with big dataset. : UbiComp 2015 - Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing (pp. 1269-1280). (UbiComp 2015 - Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing). Association for Computing Machinery, Inc. https://doi.org/10.1145/2750858.2807533

Mobile activity recognition for a whole day : Recognizing real nursing activities with big dataset. / Inoue, Sozo; Ueda, Naonori; Nohara, Yasunobu; Nakashima, Naoki.

UbiComp 2015 - Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing. Association for Computing Machinery, Inc, 2015. p. 1269-1280 (UbiComp 2015 - Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing).

研究成果: 著書/レポートタイプへの貢献会議での発言

Inoue, S, Ueda, N, Nohara, Y & Nakashima, N 2015, Mobile activity recognition for a whole day: Recognizing real nursing activities with big dataset. : UbiComp 2015 - Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing. UbiComp 2015 - Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing, Association for Computing Machinery, Inc, pp. 1269-1280, 3rd ACM International Joint Conference on Pervasive and Ubiquitous Computing, UbiComp 2015, Osaka, 日本, 9/7/15. https://doi.org/10.1145/2750858.2807533
Inoue S, Ueda N, Nohara Y, Nakashima N. Mobile activity recognition for a whole day: Recognizing real nursing activities with big dataset. : UbiComp 2015 - Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing. Association for Computing Machinery, Inc. 2015. p. 1269-1280. (UbiComp 2015 - Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing). https://doi.org/10.1145/2750858.2807533
Inoue, Sozo ; Ueda, Naonori ; Nohara, Yasunobu ; Nakashima, Naoki. / Mobile activity recognition for a whole day : Recognizing real nursing activities with big dataset. UbiComp 2015 - Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing. Association for Computing Machinery, Inc, 2015. pp. 1269-1280 (UbiComp 2015 - Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing).
@inproceedings{f4be80d9f96642c5a84df5cb6cfe9459,
title = "Mobile activity recognition for a whole day: Recognizing real nursing activities with big dataset",
abstract = "In this paper, we provide a real nursing data set for mobile activity recognition that can be used for supervised machine learning, and big data combined the patient medical records and sensors attempted for 2 years, and also propose a method for recognizing activities for a whole day utilizing prior knowledge about the activity segments in a day. Furthermore, we demonstrate data mining by applying our method to the bigger data with additional hospital data. In the proposed method, we 1) convert a set of segment timestamps into a prior probability of the activity segment by exploiting the concept of importance sampling, 2) obtain the likelihood of traditional recognition methods for each local time window within the segment range, and, 3) apply Bayesian estimation by marginalizing the conditional probability of estimating the activities for the segment samples. By evaluating with the dataset, the proposed method outperformed the traditional method without using the prior knowledge by 25.81{\%} at maximum by balanced classification rate. Moreover, the proposed method significantly reduces duration errors of activity segments from 324.2 seconds of the traditional method to 74.6 seconds at maximum. We also demonstrate the data mining by applying our method to bigger data in a hospital.",
author = "Sozo Inoue and Naonori Ueda and Yasunobu Nohara and Naoki Nakashima",
year = "2015",
month = "9",
day = "7",
doi = "10.1145/2750858.2807533",
language = "English",
series = "UbiComp 2015 - Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing",
publisher = "Association for Computing Machinery, Inc",
pages = "1269--1280",
booktitle = "UbiComp 2015 - Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing",

}

TY - GEN

T1 - Mobile activity recognition for a whole day

T2 - Recognizing real nursing activities with big dataset

AU - Inoue, Sozo

AU - Ueda, Naonori

AU - Nohara, Yasunobu

AU - Nakashima, Naoki

PY - 2015/9/7

Y1 - 2015/9/7

N2 - In this paper, we provide a real nursing data set for mobile activity recognition that can be used for supervised machine learning, and big data combined the patient medical records and sensors attempted for 2 years, and also propose a method for recognizing activities for a whole day utilizing prior knowledge about the activity segments in a day. Furthermore, we demonstrate data mining by applying our method to the bigger data with additional hospital data. In the proposed method, we 1) convert a set of segment timestamps into a prior probability of the activity segment by exploiting the concept of importance sampling, 2) obtain the likelihood of traditional recognition methods for each local time window within the segment range, and, 3) apply Bayesian estimation by marginalizing the conditional probability of estimating the activities for the segment samples. By evaluating with the dataset, the proposed method outperformed the traditional method without using the prior knowledge by 25.81% at maximum by balanced classification rate. Moreover, the proposed method significantly reduces duration errors of activity segments from 324.2 seconds of the traditional method to 74.6 seconds at maximum. We also demonstrate the data mining by applying our method to bigger data in a hospital.

AB - In this paper, we provide a real nursing data set for mobile activity recognition that can be used for supervised machine learning, and big data combined the patient medical records and sensors attempted for 2 years, and also propose a method for recognizing activities for a whole day utilizing prior knowledge about the activity segments in a day. Furthermore, we demonstrate data mining by applying our method to the bigger data with additional hospital data. In the proposed method, we 1) convert a set of segment timestamps into a prior probability of the activity segment by exploiting the concept of importance sampling, 2) obtain the likelihood of traditional recognition methods for each local time window within the segment range, and, 3) apply Bayesian estimation by marginalizing the conditional probability of estimating the activities for the segment samples. By evaluating with the dataset, the proposed method outperformed the traditional method without using the prior knowledge by 25.81% at maximum by balanced classification rate. Moreover, the proposed method significantly reduces duration errors of activity segments from 324.2 seconds of the traditional method to 74.6 seconds at maximum. We also demonstrate the data mining by applying our method to bigger data in a hospital.

UR - http://www.scopus.com/inward/record.url?scp=84960924564&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84960924564&partnerID=8YFLogxK

U2 - 10.1145/2750858.2807533

DO - 10.1145/2750858.2807533

M3 - Conference contribution

AN - SCOPUS:84960924564

T3 - UbiComp 2015 - Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing

SP - 1269

EP - 1280

BT - UbiComp 2015 - Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing

PB - Association for Computing Machinery, Inc

ER -