000166742 001__ 166742
000166742 005__ 20190812205527.0
000166742 020__ $$a978-1-4577-0653-0
000166742 02470 $$2ISI$$a000298615103008
000166742 037__ $$aCONF
000166742 245__ $$aEnsemble creation and reconfiguration for activity recognition: An information theoretic approach.
000166742 269__ $$a2011
000166742 260__ $$bIeee Service Center, 445 Hoes Lane, Po Box 1331, Piscataway, Nj 08855-1331 Usa$$c2011
000166742 336__ $$aConference Papers
000166742 490__ $$aIEEE International Conference on Systems Man and Cybernetics Conference Proceedings
000166742 520__ $$aTechnological advances in sensing and portable computing devices and wireless communication has lead to an increase in the number and variety of sensing enabled devices (e.g. smartphones or sensing garments). Pervasive computing and activity recognition systems should be able to take advantage of these sensors, even if they are not always available, or appear during the system operation. These sensors can be integrated into an ensemble where information from each sensor is then fused to obtain the system decisions. There is therefore a need for mechanisms to select which sensors should compose the ensemble, as well as techniques for dynamically reconfigure the ensemble so as to integrate new sensors. From the machine learning point of view, this approach corresponds to the combination of classifiers where measures of the accuracy and diversity of the ensemble are used to select the elements that may lead to the highest performance. Recent works have proposed measures of accuracy and diversity based on an information theoretical approach. In this paper we study the use of these measures for selecting ensembles in activity recognition based on body sensor networks. In addition to compare the obtained performance with traditional diversity measures (e.g., Q-, κ-statistics) we also present mechanisms to exploit these measures for the dynamic reconfiguration of the ensemble and detection of changes in the network (e.g. due to sensor noise or malfunction).
000166742 6531_ $$aClassifier combination
000166742 6531_ $$aMutual information
000166742 6531_ $$aDiversity
000166742 6531_ $$aDynamic reconfiguration
000166742 6531_ $$aBody sensor networks
000166742 6531_ $$a[Opportunity]
000166742 700__ $$0241256$$g137762$$aChavarriaga, Ricardo
000166742 700__ $$0242182$$g191533$$aSagha, Hesam
000166742 700__ $$aMillán, José del R.$$g149175$$0240030
000166742 7112_ $$dOctober 9-12, 2011$$cAnchorage, Alaska, USA$$a IEEE Int Conf Systems, Man, and Cybernetics (IEEE SMC 2011)
000166742 773__ $$t2011 Ieee International Conference On Systems, Man, And Cybernetics (Smc)$$q2761-2766
000166742 8564_ $$zn/a$$yn/a$$uhttps://infoscience.epfl.ch/record/166742/files/ITfusion.pdf$$s353491
000166742 909C0 $$xU12103$$pCNBI$$0252018
000166742 909C0 $$0252517$$xU12599$$pCNP
000166742 909CO $$qGLOBAL_SET$$pconf$$pSTI$$ooai:infoscience.tind.io:166742
000166742 917Z8 $$x137762
000166742 917Z8 $$x137762
000166742 917Z8 $$x137762
000166742 917Z8 $$x137762
000166742 937__ $$aEPFL-CONF-166742
000166742 973__ $$rREVIEWED$$sPUBLISHED$$aEPFL
000166742 980__ $$aCONF