Abstract

The abundance of data available on Wireless Sensor Networks makes online processing necessary. In industrial applications, for example, the correct operation of equipment can be the point of interest. The raw sampled data is of minor importance. Classification algorithms can be used to make state classifications based on the available data for devices such as industrial refrigerators. The reliability through redundancy approach used in Wireless Sensor Networks complicates practical realizations of classification algorithms. Individual inputs are susceptible to multiple disturbances like hardware failure, communication failure and battery depletion. In order to demonstrate the effects of input failure on classification algorithms, we have compared three widely used algorithms in multiple error scenarios. The compared algorithms are Feed Forward Neural Networks, naive Bayes classifiers and decision trees. Using a new experimental data-set, we show that the performance under error scenarios degrades less for the naive Bayes classifier than for the two other algorithms.
Original languageUndefined
Title of host publication8th International ICST Conference on Mobile and Ubiquitous Systems: Computing, Networking, and Services, MobiQuitous 2011
Place of PublicationLondon
PublisherSpringer Verlag
Pages126-137
Number of pages12
ISBN (Print)978-3-642-30972-4
DOIs
StatePublished - 6 Dec 2011

Publication series

NameLecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering (LNICST)
PublisherSpringer Verlag
Volume104
ISSN (Print)1867-8211

Fingerprint

Wireless sensor networks
Classifiers
Refrigerators
Feedforward neural networks
Decision trees
Industrial applications
Redundancy
Hardware
Communication

Keywords

  • METIS-285058
  • EWI-21380
  • IR-79624

Cite this

Zwartjes, G. J., Bahrepour, M., Havinga, P. J. M., Hurink, J. L., & Smit, G. J. M. (2011). On the Effects of Input Unreliability on Classifion Algorithms. In 8th International ICST Conference on Mobile and Ubiquitous Systems: Computing, Networking, and Services, MobiQuitous 2011 (pp. 126-137). (Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering (LNICST); Vol. 104). London: Springer Verlag. DOI: 10.1007/978-3-642-30973-1_11

Zwartjes, G.J.; Bahrepour, M.; Havinga, Paul J.M.; Hurink, Johann L.; Smit, Gerardus Johannes Maria / On the Effects of Input Unreliability on Classifion Algorithms.

8th International ICST Conference on Mobile and Ubiquitous Systems: Computing, Networking, and Services, MobiQuitous 2011. London : Springer Verlag, 2011. p. 126-137 (Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering (LNICST); Vol. 104).

Research output: Scientific - peer-reviewConference contribution

@inbook{2b0a97cd4da54b95ba461072d7d81a78,
title = "On the Effects of Input Unreliability on Classifion Algorithms",
abstract = "The abundance of data available on Wireless Sensor Networks makes online processing necessary. In industrial applications, for example, the correct operation of equipment can be the point of interest. The raw sampled data is of minor importance. Classification algorithms can be used to make state classifications based on the available data for devices such as industrial refrigerators. The reliability through redundancy approach used in Wireless Sensor Networks complicates practical realizations of classification algorithms. Individual inputs are susceptible to multiple disturbances like hardware failure, communication failure and battery depletion. In order to demonstrate the effects of input failure on classification algorithms, we have compared three widely used algorithms in multiple error scenarios. The compared algorithms are Feed Forward Neural Networks, naive Bayes classifiers and decision trees. Using a new experimental data-set, we show that the performance under error scenarios degrades less for the naive Bayes classifier than for the two other algorithms.",
keywords = "METIS-285058, EWI-21380, IR-79624",
author = "G.J. Zwartjes and M. Bahrepour and Havinga, {Paul J.M.} and Hurink, {Johann L.} and Smit, {Gerardus Johannes Maria}",
note = "http://mobiquitous.org/2011/",
year = "2011",
month = "12",
doi = "10.1007/978-3-642-30973-1_11",
isbn = "978-3-642-30972-4",
series = "Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering (LNICST)",
publisher = "Springer Verlag",
pages = "126--137",
booktitle = "8th International ICST Conference on Mobile and Ubiquitous Systems: Computing, Networking, and Services, MobiQuitous 2011",

}

Zwartjes, GJ, Bahrepour, M, Havinga, PJM, Hurink, JL & Smit, GJM 2011, On the Effects of Input Unreliability on Classifion Algorithms. in 8th International ICST Conference on Mobile and Ubiquitous Systems: Computing, Networking, and Services, MobiQuitous 2011. Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering (LNICST), vol. 104, Springer Verlag, London, pp. 126-137. DOI: 10.1007/978-3-642-30973-1_11

On the Effects of Input Unreliability on Classifion Algorithms. / Zwartjes, G.J.; Bahrepour, M.; Havinga, Paul J.M.; Hurink, Johann L.; Smit, Gerardus Johannes Maria.

8th International ICST Conference on Mobile and Ubiquitous Systems: Computing, Networking, and Services, MobiQuitous 2011. London : Springer Verlag, 2011. p. 126-137 (Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering (LNICST); Vol. 104).

Research output: Scientific - peer-reviewConference contribution

TY - CHAP

T1 - On the Effects of Input Unreliability on Classifion Algorithms

AU - Zwartjes,G.J.

AU - Bahrepour,M.

AU - Havinga,Paul J.M.

AU - Hurink,Johann L.

AU - Smit,Gerardus Johannes Maria

N1 - http://mobiquitous.org/2011/

PY - 2011/12/6

Y1 - 2011/12/6

N2 - The abundance of data available on Wireless Sensor Networks makes online processing necessary. In industrial applications, for example, the correct operation of equipment can be the point of interest. The raw sampled data is of minor importance. Classification algorithms can be used to make state classifications based on the available data for devices such as industrial refrigerators. The reliability through redundancy approach used in Wireless Sensor Networks complicates practical realizations of classification algorithms. Individual inputs are susceptible to multiple disturbances like hardware failure, communication failure and battery depletion. In order to demonstrate the effects of input failure on classification algorithms, we have compared three widely used algorithms in multiple error scenarios. The compared algorithms are Feed Forward Neural Networks, naive Bayes classifiers and decision trees. Using a new experimental data-set, we show that the performance under error scenarios degrades less for the naive Bayes classifier than for the two other algorithms.

AB - The abundance of data available on Wireless Sensor Networks makes online processing necessary. In industrial applications, for example, the correct operation of equipment can be the point of interest. The raw sampled data is of minor importance. Classification algorithms can be used to make state classifications based on the available data for devices such as industrial refrigerators. The reliability through redundancy approach used in Wireless Sensor Networks complicates practical realizations of classification algorithms. Individual inputs are susceptible to multiple disturbances like hardware failure, communication failure and battery depletion. In order to demonstrate the effects of input failure on classification algorithms, we have compared three widely used algorithms in multiple error scenarios. The compared algorithms are Feed Forward Neural Networks, naive Bayes classifiers and decision trees. Using a new experimental data-set, we show that the performance under error scenarios degrades less for the naive Bayes classifier than for the two other algorithms.

KW - METIS-285058

KW - EWI-21380

KW - IR-79624

U2 - 10.1007/978-3-642-30973-1_11

DO - 10.1007/978-3-642-30973-1_11

M3 - Conference contribution

SN - 978-3-642-30972-4

T3 - Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering (LNICST)

SP - 126

EP - 137

BT - 8th International ICST Conference on Mobile and Ubiquitous Systems: Computing, Networking, and Services, MobiQuitous 2011

PB - Springer Verlag

ER -

Zwartjes GJ, Bahrepour M, Havinga PJM, Hurink JL, Smit GJM. On the Effects of Input Unreliability on Classifion Algorithms. In 8th International ICST Conference on Mobile and Ubiquitous Systems: Computing, Networking, and Services, MobiQuitous 2011. London: Springer Verlag. 2011. p. 126-137. (Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering (LNICST)). Available from, DOI: 10.1007/978-3-642-30973-1_11