Original title: Klasifikace kovů pomocí spektroskopie laserem buzeného plazmatu a chemometrických metod
Translated title: Classification of metals by means of Laser-induced Breakdown Spectroscopy and chemometric methods
Authors: Képeš, Erik ; Hrdlička, Aleš (referee) ; Novotný, Jan (advisor)
Document type: Master’s theses
Year: 2017
Language: eng
Publisher: Vysoké učení technické v Brně. Fakulta strojního inženýrství
Abstract: [eng] [cze]
Táto diplomová práca sa zaoberá klasifikáciou kovov pomocou spektroskopie laserom indukovanej plazmy (LIBS) a chemometrických metód. Práca poskytuje prehľad o štúdiách na danú tému. Sú vybrané tri široko používané chemometrické klasifikačné metódy: "Soft Independent Modeling of Class Analogy" (SIMCA), "Partial Least Squares Discriminant Analysis" (PLS-DA) a variácia umelých neurónových sietí (ANN), "Feedforward Multilayer Perceptron". Rôzne prístupy k prieskumovej analýze su tiež preskúmané. Metódy sú stručne opísané. Následne sú klasifikátory experimentálne porovnané. This thesis deals with the classification of metals by means of laser-induced breakdown spectroscopy (LIBS) and chemometric methods. The work gives a review of the studies reported on the subject. Three widely used chemometric classification methods are selected: Soft Independent Modelling of Class Analogy (SIMCA), Partial Least Squares Discriminant Analysis (PLS-DA) and a variation of Artificial Neural Networks (ANN), the Feedforward Multilayer Perceptron. Several approaches to exploratory data analysis are also considered. The methods are described, briefly stating their working principle. Subsequently, the performance of the classifiers is experimentally assessed, using several figures of merit.
Keywords: ANN; chemometria; klasifikácia; LIBS; PLS-DA; SIMCA; spektroskopia laserom indukovanej plazmy; ANN; chemometrics; classification; laser-induced breakdown spectroscopy; LIBS; PLS-DA; SIMCA

Institution: Brno University of Technology (web)
Document availability information: Fulltext is available in the Brno University of Technology Digital Library.
Original record: http://hdl.handle.net/11012/66640

Permalink: http://www.nusl.cz/ntk/nusl-318788


The record appears in these collections:
Universities and colleges > Public universities > Brno University of Technology
Academic theses (ETDs) > Master’s theses