Deep neural networks for activity recognition with multi-sensor data in a smart home

Jiho Park, Kiyoung Jang, Sung-Bong Yang

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

3 Citations (Scopus)

Abstract

Multi-sensor based human activity recognition is one of the challenges in the ambient intelligent environments such as smart home and smart city. Ordinary people in their daily lives usually share a similar and repetitive life pattern, also known as life cycle. Smart home environment and its multi sensors can provide assistance to human by collecting the data sequence of human activities to predict the desired actions. Our goal is to analyze the sequence of activities recorded by a specific resident using deep learning with multiple sensor data. In this paper, we train the multiple sensor data collected by a smart home using several deep neural networks. According to the characteristics of the Recurrent Neural Network (RNN) structure, multiple sensor data of smart home is suitable for RNN because it has a sequence data in time. To support our assumption, we proposed the Residual-RNN architecture to predict future activities of a resident. Furthermore, we also utilized attention module to filter out the meaningless data to have more effective results than the one without. To verify our proposed idea, we used real resident activity in smart home using Massachusetts Institute of Technology (MIT) dataset. After our experiments, our proposed model with attention mechanism outperform the Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) model in terms of predicting the desired activities of a smart home resident.

Original languageEnglish
Title of host publicationIEEE World Forum on Internet of Things, WF-IoT 2018 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages155-160
Number of pages6
ISBN (Electronic)9781467399449
DOIs
Publication statusPublished - 2018 May 4
Event4th IEEE World Forum on Internet of Things, WF-IoT 2018 - Singapore, Singapore
Duration: 2018 Feb 52018 Feb 8

Publication series

NameIEEE World Forum on Internet of Things, WF-IoT 2018 - Proceedings
Volume2018-January

Other

Other4th IEEE World Forum on Internet of Things, WF-IoT 2018
CountrySingapore
CitySingapore
Period18/2/518/2/8

Fingerprint

Recurrent neural networks
Sensors
Network architecture
Life cycle
Deep neural networks
Sensor
Neural networks
Residents
Experiments

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Computer Networks and Communications
  • Computer Science Applications
  • Hardware and Architecture
  • Information Systems and Management
  • Safety, Risk, Reliability and Quality

Cite this

Park, J., Jang, K., & Yang, S-B. (2018). Deep neural networks for activity recognition with multi-sensor data in a smart home. In IEEE World Forum on Internet of Things, WF-IoT 2018 - Proceedings (pp. 155-160). (IEEE World Forum on Internet of Things, WF-IoT 2018 - Proceedings; Vol. 2018-January). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/WF-IoT.2018.8355147
Park, Jiho ; Jang, Kiyoung ; Yang, Sung-Bong. / Deep neural networks for activity recognition with multi-sensor data in a smart home. IEEE World Forum on Internet of Things, WF-IoT 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 155-160 (IEEE World Forum on Internet of Things, WF-IoT 2018 - Proceedings).
@inproceedings{b6bcfb05ae904ce5961dc72360a2dc68,
title = "Deep neural networks for activity recognition with multi-sensor data in a smart home",
abstract = "Multi-sensor based human activity recognition is one of the challenges in the ambient intelligent environments such as smart home and smart city. Ordinary people in their daily lives usually share a similar and repetitive life pattern, also known as life cycle. Smart home environment and its multi sensors can provide assistance to human by collecting the data sequence of human activities to predict the desired actions. Our goal is to analyze the sequence of activities recorded by a specific resident using deep learning with multiple sensor data. In this paper, we train the multiple sensor data collected by a smart home using several deep neural networks. According to the characteristics of the Recurrent Neural Network (RNN) structure, multiple sensor data of smart home is suitable for RNN because it has a sequence data in time. To support our assumption, we proposed the Residual-RNN architecture to predict future activities of a resident. Furthermore, we also utilized attention module to filter out the meaningless data to have more effective results than the one without. To verify our proposed idea, we used real resident activity in smart home using Massachusetts Institute of Technology (MIT) dataset. After our experiments, our proposed model with attention mechanism outperform the Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) model in terms of predicting the desired activities of a smart home resident.",
author = "Jiho Park and Kiyoung Jang and Sung-Bong Yang",
year = "2018",
month = "5",
day = "4",
doi = "10.1109/WF-IoT.2018.8355147",
language = "English",
series = "IEEE World Forum on Internet of Things, WF-IoT 2018 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "155--160",
booktitle = "IEEE World Forum on Internet of Things, WF-IoT 2018 - Proceedings",
address = "United States",

}

Park, J, Jang, K & Yang, S-B 2018, Deep neural networks for activity recognition with multi-sensor data in a smart home. in IEEE World Forum on Internet of Things, WF-IoT 2018 - Proceedings. IEEE World Forum on Internet of Things, WF-IoT 2018 - Proceedings, vol. 2018-January, Institute of Electrical and Electronics Engineers Inc., pp. 155-160, 4th IEEE World Forum on Internet of Things, WF-IoT 2018, Singapore, Singapore, 18/2/5. https://doi.org/10.1109/WF-IoT.2018.8355147

Deep neural networks for activity recognition with multi-sensor data in a smart home. / Park, Jiho; Jang, Kiyoung; Yang, Sung-Bong.

IEEE World Forum on Internet of Things, WF-IoT 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. p. 155-160 (IEEE World Forum on Internet of Things, WF-IoT 2018 - Proceedings; Vol. 2018-January).

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

TY - GEN

T1 - Deep neural networks for activity recognition with multi-sensor data in a smart home

AU - Park, Jiho

AU - Jang, Kiyoung

AU - Yang, Sung-Bong

PY - 2018/5/4

Y1 - 2018/5/4

N2 - Multi-sensor based human activity recognition is one of the challenges in the ambient intelligent environments such as smart home and smart city. Ordinary people in their daily lives usually share a similar and repetitive life pattern, also known as life cycle. Smart home environment and its multi sensors can provide assistance to human by collecting the data sequence of human activities to predict the desired actions. Our goal is to analyze the sequence of activities recorded by a specific resident using deep learning with multiple sensor data. In this paper, we train the multiple sensor data collected by a smart home using several deep neural networks. According to the characteristics of the Recurrent Neural Network (RNN) structure, multiple sensor data of smart home is suitable for RNN because it has a sequence data in time. To support our assumption, we proposed the Residual-RNN architecture to predict future activities of a resident. Furthermore, we also utilized attention module to filter out the meaningless data to have more effective results than the one without. To verify our proposed idea, we used real resident activity in smart home using Massachusetts Institute of Technology (MIT) dataset. After our experiments, our proposed model with attention mechanism outperform the Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) model in terms of predicting the desired activities of a smart home resident.

AB - Multi-sensor based human activity recognition is one of the challenges in the ambient intelligent environments such as smart home and smart city. Ordinary people in their daily lives usually share a similar and repetitive life pattern, also known as life cycle. Smart home environment and its multi sensors can provide assistance to human by collecting the data sequence of human activities to predict the desired actions. Our goal is to analyze the sequence of activities recorded by a specific resident using deep learning with multiple sensor data. In this paper, we train the multiple sensor data collected by a smart home using several deep neural networks. According to the characteristics of the Recurrent Neural Network (RNN) structure, multiple sensor data of smart home is suitable for RNN because it has a sequence data in time. To support our assumption, we proposed the Residual-RNN architecture to predict future activities of a resident. Furthermore, we also utilized attention module to filter out the meaningless data to have more effective results than the one without. To verify our proposed idea, we used real resident activity in smart home using Massachusetts Institute of Technology (MIT) dataset. After our experiments, our proposed model with attention mechanism outperform the Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) model in terms of predicting the desired activities of a smart home resident.

UR - http://www.scopus.com/inward/record.url?scp=85050396935&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85050396935&partnerID=8YFLogxK

U2 - 10.1109/WF-IoT.2018.8355147

DO - 10.1109/WF-IoT.2018.8355147

M3 - Conference contribution

T3 - IEEE World Forum on Internet of Things, WF-IoT 2018 - Proceedings

SP - 155

EP - 160

BT - IEEE World Forum on Internet of Things, WF-IoT 2018 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Park J, Jang K, Yang S-B. Deep neural networks for activity recognition with multi-sensor data in a smart home. In IEEE World Forum on Internet of Things, WF-IoT 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2018. p. 155-160. (IEEE World Forum on Internet of Things, WF-IoT 2018 - Proceedings). https://doi.org/10.1109/WF-IoT.2018.8355147