Abstract

Online processing is essential for many sensor network applications. Sensor nodes can sample far more data than what can practically be transmitted using state of the art sensor network radios. Online processing, however, is complicated due to limited resources of individual nodes. The naive Bayes classifier is an algorithm proven to be suitable for online classification on Wireless Sensor Networks. In this paper, we investigate a new technique to improve the naive Bayes classifier while maintaining sensor network compatibility. We propose the application of unsupervised learning techniques to enhance the probability density estimation needed for naive Bayes, thereby achieving the benefits of binning histogram probability density estimation without the related memory requirements. Using an offline experimental dataset, we demonstrate the possibility of matching the performance of the binning histogram approach within the constraints provided by Wireless Sensor Network hardware. We validate the feasibility of our approach using an implementation based on Arduino Nano hardware combined with NRF24L01+ radios.
Original languageUndefined
Title of host publicationProceedings of the Sixth International Conference on MObile Ubiquitous Computing, Systems, Services and Technologies, UBICOMM 2012
Place of PublicationUSA
PublisherXpert Publishing Services
Pages71-76
Number of pages6
ISBN (Print)978-1-61208-236-3
StatePublished - 2012

Publication series

Name
PublisherXpert Publishing Services

Fingerprint

Sensor networks
Radio receivers
Wireless sensor networks
Classifiers
Hardware
Unsupervised learning
Sensor nodes

Keywords

  • EWI-22478
  • METIS-289772
  • IR-82154

Cite this

Zwartjes, G. J., Havinga, P. J. M., Smit, G. J. M., & Hurink, J. L. (2012). Using Unsupervised Learning to Improve the Naive Bayes Classifier for Wireless Sensor Networks. In Proceedings of the Sixth International Conference on MObile Ubiquitous Computing, Systems, Services and Technologies, UBICOMM 2012 (pp. 71-76). USA: Xpert Publishing Services.

Zwartjes, G.J.; Havinga, Paul J.M.; Smit, Gerardus Johannes Maria; Hurink, Johann L. / Using Unsupervised Learning to Improve the Naive Bayes Classifier for Wireless Sensor Networks.

Proceedings of the Sixth International Conference on MObile Ubiquitous Computing, Systems, Services and Technologies, UBICOMM 2012. USA : Xpert Publishing Services, 2012. p. 71-76.

Research output: Scientific - peer-reviewConference contribution

@inbook{630d7bd66b4247dc96e8ffa9b4f0d6d0,
title = "Using Unsupervised Learning to Improve the Naive Bayes Classifier for Wireless Sensor Networks",
abstract = "Online processing is essential for many sensor network applications. Sensor nodes can sample far more data than what can practically be transmitted using state of the art sensor network radios. Online processing, however, is complicated due to limited resources of individual nodes. The naive Bayes classifier is an algorithm proven to be suitable for online classification on Wireless Sensor Networks. In this paper, we investigate a new technique to improve the naive Bayes classifier while maintaining sensor network compatibility. We propose the application of unsupervised learning techniques to enhance the probability density estimation needed for naive Bayes, thereby achieving the benefits of binning histogram probability density estimation without the related memory requirements. Using an offline experimental dataset, we demonstrate the possibility of matching the performance of the binning histogram approach within the constraints provided by Wireless Sensor Network hardware. We validate the feasibility of our approach using an implementation based on Arduino Nano hardware combined with NRF24L01+ radios.",
keywords = "EWI-22478, METIS-289772, IR-82154",
author = "G.J. Zwartjes and Havinga, {Paul J.M.} and Smit, {Gerardus Johannes Maria} and Hurink, {Johann L.}",
year = "2012",
isbn = "978-1-61208-236-3",
publisher = "Xpert Publishing Services",
pages = "71--76",
booktitle = "Proceedings of the Sixth International Conference on MObile Ubiquitous Computing, Systems, Services and Technologies, UBICOMM 2012",

}

Zwartjes, GJ, Havinga, PJM, Smit, GJM & Hurink, JL 2012, Using Unsupervised Learning to Improve the Naive Bayes Classifier for Wireless Sensor Networks. in Proceedings of the Sixth International Conference on MObile Ubiquitous Computing, Systems, Services and Technologies, UBICOMM 2012. Xpert Publishing Services, USA, pp. 71-76.

Using Unsupervised Learning to Improve the Naive Bayes Classifier for Wireless Sensor Networks. / Zwartjes, G.J.; Havinga, Paul J.M.; Smit, Gerardus Johannes Maria; Hurink, Johann L.

Proceedings of the Sixth International Conference on MObile Ubiquitous Computing, Systems, Services and Technologies, UBICOMM 2012. USA : Xpert Publishing Services, 2012. p. 71-76.

Research output: Scientific - peer-reviewConference contribution

TY - CHAP

T1 - Using Unsupervised Learning to Improve the Naive Bayes Classifier for Wireless Sensor Networks

AU - Zwartjes,G.J.

AU - Havinga,Paul J.M.

AU - Smit,Gerardus Johannes Maria

AU - Hurink,Johann L.

PY - 2012

Y1 - 2012

N2 - Online processing is essential for many sensor network applications. Sensor nodes can sample far more data than what can practically be transmitted using state of the art sensor network radios. Online processing, however, is complicated due to limited resources of individual nodes. The naive Bayes classifier is an algorithm proven to be suitable for online classification on Wireless Sensor Networks. In this paper, we investigate a new technique to improve the naive Bayes classifier while maintaining sensor network compatibility. We propose the application of unsupervised learning techniques to enhance the probability density estimation needed for naive Bayes, thereby achieving the benefits of binning histogram probability density estimation without the related memory requirements. Using an offline experimental dataset, we demonstrate the possibility of matching the performance of the binning histogram approach within the constraints provided by Wireless Sensor Network hardware. We validate the feasibility of our approach using an implementation based on Arduino Nano hardware combined with NRF24L01+ radios.

AB - Online processing is essential for many sensor network applications. Sensor nodes can sample far more data than what can practically be transmitted using state of the art sensor network radios. Online processing, however, is complicated due to limited resources of individual nodes. The naive Bayes classifier is an algorithm proven to be suitable for online classification on Wireless Sensor Networks. In this paper, we investigate a new technique to improve the naive Bayes classifier while maintaining sensor network compatibility. We propose the application of unsupervised learning techniques to enhance the probability density estimation needed for naive Bayes, thereby achieving the benefits of binning histogram probability density estimation without the related memory requirements. Using an offline experimental dataset, we demonstrate the possibility of matching the performance of the binning histogram approach within the constraints provided by Wireless Sensor Network hardware. We validate the feasibility of our approach using an implementation based on Arduino Nano hardware combined with NRF24L01+ radios.

KW - EWI-22478

KW - METIS-289772

KW - IR-82154

M3 - Conference contribution

SN - 978-1-61208-236-3

SP - 71

EP - 76

BT - Proceedings of the Sixth International Conference on MObile Ubiquitous Computing, Systems, Services and Technologies, UBICOMM 2012

PB - Xpert Publishing Services

ER -

Zwartjes GJ, Havinga PJM, Smit GJM, Hurink JL. Using Unsupervised Learning to Improve the Naive Bayes Classifier for Wireless Sensor Networks. In Proceedings of the Sixth International Conference on MObile Ubiquitous Computing, Systems, Services and Technologies, UBICOMM 2012. USA: Xpert Publishing Services. 2012. p. 71-76.