Algorithms for enhancing pattern separability, feature selection and incremental learning with applications to gas sensing electronic nose systems

منو

" Algorithms for enhancing pattern separability, feature selection and incremental learning with applications to gas sensing electronic nose systems "

Document Type	:	Latin Dissertation
Language of Document	:	English
Record Number	:	151519
Doc. No	:	ET23311
Main Entry	:	Robi Polikslr
Title Proper	:	Algorithms for enhancing pattern separability, feature selection and incremental learning with applications to gas sensing electronic nose systems
Note	:	This document is digital این مدرک بصورت الکترونیکی می باشد
Abstract	:	Three major issues in pattern recognition and data analysis have been addressed in thisstudy and applied to the problem of identification of volatile organic compounds (VOC) forgas sensing applications. Various approaches have been proposed and discussed. These ap-proaches are not only applicable to the VOC identification. but also to a variety of patternrecognition and data analysis problems. In particular,enhancing pat tern separability for challenging classifcat ion pro blerns.optimum feature selection problem, andincremental leaming for neural networkshave been investigated.Three different approaches are proposed for enhancing pattern separability for classifica-tion of closely spaced. or possibly overlapping clusters. In the neurohzzy approach. a fuzzyinference system that considers the dynamic ranges of individual features is developed. Fea-ture range stretching (FRS) is introduced as an alternative approach for increasing interclus-ter distances by mapping the tight dynamic range of each feature to a wider through anonlinear function. Finally. a third approach, nonlinear cluster transformation....-....,....-...,..tested for theQ1 PC1 bus cardBoth these projects mere sofixare des elopment efforts tonards contributing to dlfferentaspects of Roboucs and lZ1echatronics projects m the Controls and Roboucs Group..
Subject	:	Electericl tess
	:	برق
electronic file name	:	TL46548.pdf
Title and statement of responsibility and	:	Algorithms for enhancing pattern separability, feature selection and incremental learning with applications to gas sensing electronic nose systems [Thesis]