Title :
Data Uncertainty Sensitivity Analysis for Reduced Complexity SVM Classifiers
Author :
Gubian, M. ; Boni, A. ; Petri, D.
Author_Institution :
Dept. of Inf. & Commun. Technol., Trento Univ.
Abstract :
In this paper we investigate experimentally how different sources of uncertainty affect the classification performance of an SVM based binary classifier. Our aim is to find statistically sound methods for controlling the detrimental effects of such sources when a classifier is to be implemented in hardware platforms where severe limitations force designers to allocate power, computation and memory resources carefully. At a first analysis, SVM revealed robust in terms of noise on data, whereas training data scarcity is a problem to be investigated further on
Keywords :
intelligent sensors; measurement uncertainty; signal classification; support vector machines; wireless sensor networks; SVM classifiers; binary classifier; data uncertainty sensitivity analysis; smart sensors; support vector machines; Acoustic noise; Force control; Hardware; Noise robustness; Resource management; Sensitivity analysis; Support vector machine classification; Support vector machines; Training data; Uncertainty; Support Vector Machines (SVMs); model selection; noise robustness; smart sensors;
Conference_Titel :
Instrumentation and Measurement Technology Conference, 2006. IMTC 2006. Proceedings of the IEEE
Conference_Location :
Sorrento
Print_ISBN :
0-7803-9359-7
Electronic_ISBN :
1091-5281
DOI :
10.1109/IMTC.2006.328647
