Multi-stage activity inference for locomotion and transportation analytics of mobile users

Yugo Nakamura, Yoshinori Umetsu, Jose Paolo Talusan, Keiichi Yasumoto, Wataru Sasaki, Masashi Takata, Yutaka Arakawa

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

In this paper, we, Ubi-NUTS Japan, introduce a multi-stage activity inference method that can recognize a user's mode of locomotion and transportation using mobile device sensors. We use the Sussex-Huawei Locomotion-Transportation (SHL) dataset to tackle the SHL recognition challenge, where the goal is to recognize 8 modes of locomotion and transportation (still, walk, run, bike, car, bus, train, and subway) activities from the inertial sensor data of a smartphone. We adopt a multi-stage approach where the 8 class classification problem is divided into multiple sub-problems considering the similarity of each activity. Multimodal sensor data collected from a mobile phone are inferred using a proposed pipeline that combines feature extraction and 4 different types of classifiers generated using the random forest algorithm. We evaluated our method using data from over 271 hours of daily activities of 1 participant and the 5-fold cross-validation. Evaluation results demonstrate that our method clearly recognizes the 8 types of activities with an average F1-score of 97%.

Original languageEnglish
Title of host publicationUbiComp/ISWC 2018 - Adjunct Proceedings of the 2018 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2018 ACM International Symposium on Wearable Computers
PublisherAssociation for Computing Machinery, Inc
Pages1579-1588
Number of pages10
ISBN (Electronic)9781450359665
DOIs
Publication statusPublished - Oct 8 2018
Externally publishedYes
Event2018 Joint ACM International Conference on Pervasive and Ubiquitous Computing, UbiComp 2018 and 2018 ACM International Symposium on Wearable Computers, ISWC 2018 - Singapore, Singapore
Duration: Oct 8 2018Oct 12 2018

Publication series

NameUbiComp/ISWC 2018 - Adjunct Proceedings of the 2018 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2018 ACM International Symposium on Wearable Computers

Other

Other2018 Joint ACM International Conference on Pervasive and Ubiquitous Computing, UbiComp 2018 and 2018 ACM International Symposium on Wearable Computers, ISWC 2018
CountrySingapore
CitySingapore
Period10/8/1810/12/18

Fingerprint

Sensors
Subways
Smartphones
Mobile phones
Mobile devices
Feature extraction
Classifiers
Railroad cars
Pipelines

All Science Journal Classification (ASJC) codes

  • Software
  • Human-Computer Interaction
  • Information Systems

Cite this

Nakamura, Y., Umetsu, Y., Talusan, J. P., Yasumoto, K., Sasaki, W., Takata, M., & Arakawa, Y. (2018). Multi-stage activity inference for locomotion and transportation analytics of mobile users. In UbiComp/ISWC 2018 - Adjunct Proceedings of the 2018 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2018 ACM International Symposium on Wearable Computers (pp. 1579-1588). (UbiComp/ISWC 2018 - Adjunct Proceedings of the 2018 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2018 ACM International Symposium on Wearable Computers). Association for Computing Machinery, Inc. https://doi.org/10.1145/3267305.3267526

Multi-stage activity inference for locomotion and transportation analytics of mobile users. / Nakamura, Yugo; Umetsu, Yoshinori; Talusan, Jose Paolo; Yasumoto, Keiichi; Sasaki, Wataru; Takata, Masashi; Arakawa, Yutaka.

UbiComp/ISWC 2018 - Adjunct Proceedings of the 2018 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2018 ACM International Symposium on Wearable Computers. Association for Computing Machinery, Inc, 2018. p. 1579-1588 (UbiComp/ISWC 2018 - Adjunct Proceedings of the 2018 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2018 ACM International Symposium on Wearable Computers).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Nakamura, Y, Umetsu, Y, Talusan, JP, Yasumoto, K, Sasaki, W, Takata, M & Arakawa, Y 2018, Multi-stage activity inference for locomotion and transportation analytics of mobile users. in UbiComp/ISWC 2018 - Adjunct Proceedings of the 2018 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2018 ACM International Symposium on Wearable Computers. UbiComp/ISWC 2018 - Adjunct Proceedings of the 2018 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2018 ACM International Symposium on Wearable Computers, Association for Computing Machinery, Inc, pp. 1579-1588, 2018 Joint ACM International Conference on Pervasive and Ubiquitous Computing, UbiComp 2018 and 2018 ACM International Symposium on Wearable Computers, ISWC 2018, Singapore, Singapore, 10/8/18. https://doi.org/10.1145/3267305.3267526
Nakamura Y, Umetsu Y, Talusan JP, Yasumoto K, Sasaki W, Takata M et al. Multi-stage activity inference for locomotion and transportation analytics of mobile users. In UbiComp/ISWC 2018 - Adjunct Proceedings of the 2018 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2018 ACM International Symposium on Wearable Computers. Association for Computing Machinery, Inc. 2018. p. 1579-1588. (UbiComp/ISWC 2018 - Adjunct Proceedings of the 2018 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2018 ACM International Symposium on Wearable Computers). https://doi.org/10.1145/3267305.3267526
Nakamura, Yugo ; Umetsu, Yoshinori ; Talusan, Jose Paolo ; Yasumoto, Keiichi ; Sasaki, Wataru ; Takata, Masashi ; Arakawa, Yutaka. / Multi-stage activity inference for locomotion and transportation analytics of mobile users. UbiComp/ISWC 2018 - Adjunct Proceedings of the 2018 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2018 ACM International Symposium on Wearable Computers. Association for Computing Machinery, Inc, 2018. pp. 1579-1588 (UbiComp/ISWC 2018 - Adjunct Proceedings of the 2018 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2018 ACM International Symposium on Wearable Computers).
@inproceedings{df98aa16981541eea82d5fc77ecef73a,
title = "Multi-stage activity inference for locomotion and transportation analytics of mobile users",
abstract = "In this paper, we, Ubi-NUTS Japan, introduce a multi-stage activity inference method that can recognize a user's mode of locomotion and transportation using mobile device sensors. We use the Sussex-Huawei Locomotion-Transportation (SHL) dataset to tackle the SHL recognition challenge, where the goal is to recognize 8 modes of locomotion and transportation (still, walk, run, bike, car, bus, train, and subway) activities from the inertial sensor data of a smartphone. We adopt a multi-stage approach where the 8 class classification problem is divided into multiple sub-problems considering the similarity of each activity. Multimodal sensor data collected from a mobile phone are inferred using a proposed pipeline that combines feature extraction and 4 different types of classifiers generated using the random forest algorithm. We evaluated our method using data from over 271 hours of daily activities of 1 participant and the 5-fold cross-validation. Evaluation results demonstrate that our method clearly recognizes the 8 types of activities with an average F1-score of 97{\%}.",
author = "Yugo Nakamura and Yoshinori Umetsu and Talusan, {Jose Paolo} and Keiichi Yasumoto and Wataru Sasaki and Masashi Takata and Yutaka Arakawa",
year = "2018",
month = "10",
day = "8",
doi = "10.1145/3267305.3267526",
language = "English",
series = "UbiComp/ISWC 2018 - Adjunct Proceedings of the 2018 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2018 ACM International Symposium on Wearable Computers",
publisher = "Association for Computing Machinery, Inc",
pages = "1579--1588",
booktitle = "UbiComp/ISWC 2018 - Adjunct Proceedings of the 2018 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2018 ACM International Symposium on Wearable Computers",

}

TY - GEN

T1 - Multi-stage activity inference for locomotion and transportation analytics of mobile users

AU - Nakamura, Yugo

AU - Umetsu, Yoshinori

AU - Talusan, Jose Paolo

AU - Yasumoto, Keiichi

AU - Sasaki, Wataru

AU - Takata, Masashi

AU - Arakawa, Yutaka

PY - 2018/10/8

Y1 - 2018/10/8

N2 - In this paper, we, Ubi-NUTS Japan, introduce a multi-stage activity inference method that can recognize a user's mode of locomotion and transportation using mobile device sensors. We use the Sussex-Huawei Locomotion-Transportation (SHL) dataset to tackle the SHL recognition challenge, where the goal is to recognize 8 modes of locomotion and transportation (still, walk, run, bike, car, bus, train, and subway) activities from the inertial sensor data of a smartphone. We adopt a multi-stage approach where the 8 class classification problem is divided into multiple sub-problems considering the similarity of each activity. Multimodal sensor data collected from a mobile phone are inferred using a proposed pipeline that combines feature extraction and 4 different types of classifiers generated using the random forest algorithm. We evaluated our method using data from over 271 hours of daily activities of 1 participant and the 5-fold cross-validation. Evaluation results demonstrate that our method clearly recognizes the 8 types of activities with an average F1-score of 97%.

AB - In this paper, we, Ubi-NUTS Japan, introduce a multi-stage activity inference method that can recognize a user's mode of locomotion and transportation using mobile device sensors. We use the Sussex-Huawei Locomotion-Transportation (SHL) dataset to tackle the SHL recognition challenge, where the goal is to recognize 8 modes of locomotion and transportation (still, walk, run, bike, car, bus, train, and subway) activities from the inertial sensor data of a smartphone. We adopt a multi-stage approach where the 8 class classification problem is divided into multiple sub-problems considering the similarity of each activity. Multimodal sensor data collected from a mobile phone are inferred using a proposed pipeline that combines feature extraction and 4 different types of classifiers generated using the random forest algorithm. We evaluated our method using data from over 271 hours of daily activities of 1 participant and the 5-fold cross-validation. Evaluation results demonstrate that our method clearly recognizes the 8 types of activities with an average F1-score of 97%.

UR - http://www.scopus.com/inward/record.url?scp=85058303103&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85058303103&partnerID=8YFLogxK

U2 - 10.1145/3267305.3267526

DO - 10.1145/3267305.3267526

M3 - Conference contribution

AN - SCOPUS:85058303103

T3 - UbiComp/ISWC 2018 - Adjunct Proceedings of the 2018 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2018 ACM International Symposium on Wearable Computers

SP - 1579

EP - 1588

BT - UbiComp/ISWC 2018 - Adjunct Proceedings of the 2018 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2018 ACM International Symposium on Wearable Computers

PB - Association for Computing Machinery, Inc

ER -