Interpretation of analytical chemical information by pattern recognition methods—a survey

Kryger, Lars

doi:10.1016/0039-9140(81)80223-8

Cited by 51 publications

(11 citation statements)

References 89 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Supervised learning techniques are used either for developing classification rules which accurately predict the classification of unknown patterns or samples (Kryger, 1981) or for finding calibration relationships between one set of measurements which are easy or cheap to acquire, and other measurements which are expensive or labour intensive, in order to predict these later (Naes et al, 2004). The simplest calibration problem consists of predicting a single response (y-variable) from a known predictor (x-variable) and can be solved by using ordinary linear regression (OLR).…”

Section: About the Data Setmentioning

confidence: 99%

Critical Aspects of Supervised Pattern Recognition Methods for Interpreting Compositional Data

González¹

2012

Chemometrics in Practical Applications

View full text Add to dashboard Cite

Section: About the Data Setmentioning

confidence: 99%

Critical Aspects of Supervised Pattern Recognition Methods for Interpreting Compositional Data

González¹

2012

Chemometrics in Practical Applications

View full text Add to dashboard Cite

“…PCA is a pattern recognition method [26,27], which allows reduction of the dimensionality of a data set by transforming the original measurement variables into new variables called principle components [26,27] arranged in the order of decreasing information content or variance. In our case, we use two first principal components to map the objects into two-dimensional space, where the clusters of proteinaceous binders can be visually observed (Figure 4).…”

Section: Analytical Proceduresmentioning

confidence: 99%

Impact of biological factors on binding media identification in art objects: Identification of animal glue in the presence of Aspergillus niger

Tsakalof

Bairachtari

Aslani

et al. 2004

J of Separation Science

View full text Add to dashboard Cite

The materials and especially organic materials used for creation of art objects can be utilized by various microorganisms for their growth and facilitate the microbial colonization of the object. An understanding of the chemical alterations in artefacts caused by the presence of microorganisms can be crucial for correct identification of the materials initially used for the artefact creation--nowadays an important step in restoration and/or art-historical investigation of the art object. The present article describes a model experiment in which we investigated the possible chemical alterations in animal glue films used as substrate for growth of the fungus Aspergillus niger. The sterilized animal glue solution was poured into Petri dishes, inoculated with Aspergillus niger, and subsequently incubated at 15 degrees C for 0, 7, 9, 14, and 28 days. After interruption of incubation, the content of the Petri dish was analyzed for amino acid composition by the GC-MS based method. It was found that the growth of Aspergillus niger on animal glue films did not cause significant changes in the amino acid composition of the film and had no impact on animal glue identification.

show abstract

“…It gives a sequence of variables on the basis of their information content (i.e., on the basis of their ability to discriminate objects belonging to different classes). The method used here classifies variables on the basis of the corresponding Fisher weight (or F-ratio) (25). The value of the Fisher weight for each variable is the average of the values of all possible pairs of categories into which the objects have been subdivided.…”

Section: Statistical Treatment Of the Datamentioning

confidence: 99%

Proton dissociation of aqueous organic acids studied by multivariate chemometrics

Aruga¹

1995

Can. J. Chem.

View full text Add to dashboard Cite

Thermodynamic data of proton dissociation of 75 organic acids belonging to four classes (protonated amines, aliphatic carboxylic acids, benzoic acids, phenols) have been processed by multivariate chemometric techniques. The variables consist of conventional thermodynamic quantities (Gibbs function, enthalpy, entropy) and of partial components of these quantities (internal and external, electrostatic and nonelectrostatic components). The above data refer to the aqueous medium, at 25OC and I = 0 mol dm-3. The Gibbs function of deprotonation in the gas phase and the Hammett a constant have also been considered. Multivariate techniques include Principal Component Analysis, Factor Analysis, and feature selection. Factor Analysis and related concepts have proved to be useful in defining the causes of differences in acid strengths and their respective importance.

show abstract

Interpretation of analytical chemical information by pattern recognition methods—a survey

Cited by 51 publications

References 89 publications

Critical Aspects of Supervised Pattern Recognition Methods for Interpreting Compositional Data

Critical Aspects of Supervised Pattern Recognition Methods for Interpreting Compositional Data

Impact of biological factors on binding media identification in art objects: Identification of animal glue in the presence of Aspergillus niger

Proton dissociation of aqueous organic acids studied by multivariate chemometrics

Contact Info

Product

Resources

About