Real-time Recognition of Interleaved Activities based on Ensemble of LSTM and Fuzzy Temporal Windows

Javier Medina Quero; Claire Orr; Shuai Zhang; CD Nugent; Alberto Salguero; Macarena Espinilla

doi:10.3390/proceedings2191225

Real-time Recognition of Interleaved Activities based on Ensemble of LSTM and Fuzzy Temporal Windows

Javier Medina Quero
, Claire Orr
, Shuai Zhang
, CD Nugent
, Alberto Salguero
, Macarena Espinilla

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

107 Downloads (Pure)

Abstract

In this paper, we present a methodology for Real-Time Activity Recognition of Interleaved Activities based on Fuzzy Logic and Recurrent Neural Networks. Firstly, we propose a representation of binary-sensor activations based on multiple Fuzzy Temporal Windows. Secondly, an ensemble of activity-based classifiers for balanced training and selection of relevant sensors is proposed. Each classifier is configured as a Long Short-Term Memory with self-reliant detection of interleaved activities. The proposed approach was evaluated using well-known interleaved binary-sensor
datasets comprised of activities of daily living.

Original language	English
Title of host publication	MDPI Proceedings
Publisher	MDPI
Number of pages	10
DOIs	https://doi.org/10.3390/proceedings2191225
Publication status	Accepted - 27 Jul 2018
Event	12h International Conference on Ubiquitous Computing and Ambient Intelligence (UCAmI) 2018 - Punta Cana, Dominican Republic Duration: 4 Dec 2018 → 7 Dec 2018

Publication series

Name	The 12th International Conference on Ubiquitous Computing and Ambient ‪Intelligence (UCAmI 2018)
Publisher	MDPI AG, Basel, Switzerland
ISSN (Electronic)	2504-3900

Conference

Conference	12h International Conference on Ubiquitous Computing and Ambient Intelligence (UCAmI) 2018
Abbreviated title	UCAmI
Country/Territory	Dominican Republic
City	Punta Cana
Period	4/12/18 → 7/12/18

Keywords

Real-Time Activity Recognition
Interleaved Activities
Fuzzy TemporalWindows
Long Short-Term Memory

Access to Document

10.3390/proceedings2191225

ucami2018_paper_64Author Accepted Manuscript, 464 KB

Cite this

Medina Quero, J., Orr, C., Zhang, S., Nugent, CD., Salguero, A., & Espinilla, M. (Accepted/In press). Real-time Recognition of Interleaved Activities based on Ensemble of LSTM and Fuzzy Temporal Windows. In MDPI Proceedings (The 12th International Conference on Ubiquitous Computing and Ambient ‪Intelligence (UCAmI 2018)). MDPI. https://doi.org/10.3390/proceedings2191225

@inproceedings{929c17dec00b4b3cb844d7e87c044c47,

title = "Real-time Recognition of Interleaved Activities based on Ensemble of LSTM and Fuzzy Temporal Windows",

abstract = "In this paper, we present a methodology for Real-Time Activity Recognition of Interleaved Activities based on Fuzzy Logic and Recurrent Neural Networks. Firstly, we propose a representation of binary-sensor activations based on multiple Fuzzy Temporal Windows. Secondly, an ensemble of activity-based classifiers for balanced training and selection of relevant sensors is proposed. Each classifier is configured as a Long Short-Term Memory with self-reliant detection of interleaved activities. The proposed approach was evaluated using well-known interleaved binary-sensor datasets comprised of activities of daily living.",

keywords = "Real-Time Activity Recognition, Interleaved Activities, Fuzzy TemporalWindows, Long Short-Term Memory",

author = "\{Medina Quero\}, Javier and Claire Orr and Shuai Zhang and CD Nugent and Alberto Salguero and Macarena Espinilla",

year = "2018",

month = jul,

day = "27",

doi = "10.3390/proceedings2191225",

language = "English",

series = "The 12th International Conference on Ubiquitous Computing and Ambient ‪Intelligence (UCAmI 2018)",

publisher = "MDPI",

booktitle = "MDPI Proceedings",

address = "Switzerland",

note = "12h International Conference on Ubiquitous Computing and Ambient Intelligence (UCAmI) 2018, UCAmI ; Conference date: 04-12-2018 Through 07-12-2018",

}

Medina Quero, J, Orr, C, Zhang, S , Nugent, CD, Salguero, A & Espinilla, M 2018, Real-time Recognition of Interleaved Activities based on Ensemble of LSTM and Fuzzy Temporal Windows. in MDPI Proceedings. The 12th International Conference on Ubiquitous Computing and Ambient ‪Intelligence (UCAmI 2018), MDPI, 12h International Conference on Ubiquitous Computing and Ambient Intelligence (UCAmI) 2018, Punta Cana, Dominican Republic, 4/12/18. https://doi.org/10.3390/proceedings2191225

Real-time Recognition of Interleaved Activities based on Ensemble of LSTM and Fuzzy Temporal Windows. / Medina Quero, Javier; Orr, Claire; Zhang, Shuai et al.
MDPI Proceedings. MDPI, 2018. (The 12th International Conference on Ubiquitous Computing and Ambient ‪Intelligence (UCAmI 2018)).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Real-time Recognition of Interleaved Activities based on Ensemble of LSTM and Fuzzy Temporal Windows

AU - Medina Quero, Javier

AU - Orr, Claire

AU - Zhang, Shuai

AU - Nugent, CD

AU - Salguero, Alberto

AU - Espinilla, Macarena

PY - 2018/7/27

Y1 - 2018/7/27

N2 - In this paper, we present a methodology for Real-Time Activity Recognition of Interleaved Activities based on Fuzzy Logic and Recurrent Neural Networks. Firstly, we propose a representation of binary-sensor activations based on multiple Fuzzy Temporal Windows. Secondly, an ensemble of activity-based classifiers for balanced training and selection of relevant sensors is proposed. Each classifier is configured as a Long Short-Term Memory with self-reliant detection of interleaved activities. The proposed approach was evaluated using well-known interleaved binary-sensor datasets comprised of activities of daily living.

AB - In this paper, we present a methodology for Real-Time Activity Recognition of Interleaved Activities based on Fuzzy Logic and Recurrent Neural Networks. Firstly, we propose a representation of binary-sensor activations based on multiple Fuzzy Temporal Windows. Secondly, an ensemble of activity-based classifiers for balanced training and selection of relevant sensors is proposed. Each classifier is configured as a Long Short-Term Memory with self-reliant detection of interleaved activities. The proposed approach was evaluated using well-known interleaved binary-sensor datasets comprised of activities of daily living.

KW - Real-Time Activity Recognition

KW - Interleaved Activities

KW - Fuzzy TemporalWindows

KW - Long Short-Term Memory

U2 - 10.3390/proceedings2191225

DO - 10.3390/proceedings2191225

M3 - Conference contribution

T3 - The 12th International Conference on Ubiquitous Computing and Ambient ‪Intelligence (UCAmI 2018)

BT - MDPI Proceedings

PB - MDPI

T2 - 12h International Conference on Ubiquitous Computing and Ambient Intelligence (UCAmI) 2018

Y2 - 4 December 2018 through 7 December 2018

ER -

Real-time Recognition of Interleaved Activities based on Ensemble of LSTM and Fuzzy Temporal Windows

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this