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 proceedingConference contribution

10 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 languageEnglish
Title of host publicationMDPI Proceedings
PublisherMDPI
Number of pages10
DOIs
Publication statusAccepted/In press - 27 Jul 2018
Event12h International Conference on Ubiquitous Computing and Ambient Intelligence (UCAmI) 2018 - Punta Cana, Dominican Republic
Duration: 4 Dec 20187 Dec 2018

Publication series

NameThe 12th International Conference on Ubiquitous Computing and Ambient ‪Intelligence (UCAmI 2018)
PublisherMDPI AG, Basel, Switzerland
ISSN (Electronic)2504-3900

Conference

Conference12h International Conference on Ubiquitous Computing and Ambient Intelligence (UCAmI) 2018
Abbreviated titleUCAmI
CountryDominican Republic
CityPunta Cana
Period4/12/187/12/18

    Fingerprint

Keywords

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

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