Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
A new approach based on temporal sub-windows for online sensor-based activity recognition
Department of Computer Science, University of Jaén, Jaén, Spain.
Department of Computer Science, University of Jaén, Jaén, Spain.
Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Computer Science.ORCID iD: 0000-0003-3191-8335
School of Computing and Mathematics, Ulster University, Coleraine, UK.
2018 (English)In: Journal of Ambient Intelligence and Humanized Computing, ISSN 1868-5137, E-ISSN 1868-5145Article in journal (Refereed) Epub ahead of print
Abstract [en]

Usually, approaches driven by data proposed in literature for sensor-based activity recognition use the begin label and the end label of each activity in the dataset, fixing a temporal window with sensor data events to identify the activity carried out in this window. This type of approach cannot be carried out in real time because it is not possible to predict the start time of an activity, i.e., the class of the future activity that an inhabitant will perform, neither when he/she will begin to carry out this activity. However, an activity can be marked as finished in real time only with the previous observations. Therefore, there is a need of online activity recognition approaches that classify activities using only the end label of the activity. In this paper, we propose and evaluate a new approach for online activity recognition with three temporal sub-windows that uses only the end label of the activity. The advantage of our approach is that the temporal sub-windows keep a partial order in the sensor data stream from the end time of the activity in a short-term, medium-term, long-term. The experiments conducted to evaluate our approach suggest the importance of the use of temporal sub-windows versus a single temporal window in terms of accuracy, using only the end time of the activity. The use of temporal sub-windows has improved the accuracy in the 98.95% of experiments carried out.

Place, publisher, year, edition, pages
Springer, 2018.
Keywords [en]
Activity recognition, Data sensor stream, Fuzzy linguistic modelling, Sensor data stream processing, Smart environments
National Category
Media and Communication Technology
Research subject
Mobile and Pervasive Computing
Identifiers
URN: urn:nbn:se:ltu:diva-67907DOI: 10.1007/s12652-018-0746-yOAI: oai:DiVA.org:ltu-67907DiVA, id: diva2:1189482
Available from: 2018-03-11 Created: 2018-03-11 Last updated: 2018-03-11

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full text

Search in DiVA

By author/editor
Hallberg, Josef
By organisation
Computer Science
In the same journal
Journal of Ambient Intelligence and Humanized Computing
Media and Communication Technology

Search outside of DiVA

GoogleGoogle Scholar

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 17 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf