Search results for "linear discriminant analysis"
showing 10 items of 163 documents
Use of linear discriminant analysis applied to vibrational spectroscopy data to characterize commercial varnishes employed for art purposes.
2007
An improvement of methodologies for characterising synthetic resins used in varnishes employed for art purposes has been suggested. Several kinds of standard of the most common polymeric resins (acrylic, vinyl, poly(vinyl alcohol), alkyd, cellulose nitrate, latex, polyester, polyurethane, epoxy, organosilicic, and ketonic) were analyzed by Fourier transform infrared (FTIR) spectroscopy. Synthetic resins characterization is based on the mathematical treatment of their whole spectrum, dividing it in 13 sections, avoiding the one-by-one interpretation of the absorption bands. The mathematical model takes as variables the maximal absorbance of each section, and each synthetic standard resin as …
Evaluation of the effect of chance correlations on variable selection using Partial Least Squares -Discriminant Analysis
2013
Variable subset selection is often mandatory in high throughput metabolomics and proteomics. However, depending on the variable to sample ratio there is a significant susceptibility of variable selection towards chance correlations. The evaluation of the predictive capabilities of PLSDA models estimated by cross-validation after feature selection provides overly optimistic results if the selection is performed on the entire set and no external validation set is available. In this work, a simulation of the statistical null hypothesis is proposed to test whether the discrimination capability of a PLSDA model after variable selection estimated by cross-validation is statistically higher than t…
Burned bones forensic investigations employing near infrared spectroscopy
2017
The use of near infrared (NIR) spectroscopy was evaluated, by using chemometric tools, for the study of the environmental impact on burned bones. Spectra of internal and external parts of burned bones, together with sediment samples, were treated by Principal Component Analysis and cluster classification as exploratory techniques to select burned bone samples, less affected by environmental processes, to properly carry out forensic studies. Partial Least Square Discriminant Analysis was used to build a model to classify bone samples based on their burning conditions, providing an efficient and accurate method to discern calcined and carbonized bone. Additionally, Partial Least Square regres…
Comparative Study of Several Machine Learning Algorithms for Classification of Unifloral Honeys
2021
Unifloral honeys are highly demanded by honey consumers, especially in Europe. To ensure that a honey belongs to a very appreciated botanical class, the classical methodology is palynological analysis to identify and count pollen grains. Highly trained personnel are needed to perform this task, which complicates the characterization of honey botanical origins. Organoleptic assessment of honey by expert personnel helps to confirm such classification. In this study, the ability of different machine learning (ML) algorithms to correctly classify seven types of Spanish honeys of single botanical origins (rosemary, citrus, lavender, sunflower, eucalyptus, heather and forest honeydew) was investi…
Use of electronic nose to determine defect percentage in oils. Comparison with sensory panel results
2010
Abstract An electronic nose based on an array of 6 metal oxide semiconductor sensors was used, jointly with linear discriminant analysis (LDA) and artificial neural network (ANN) method, to classify oils containing the five typical virgin olive oil (VOO) sensory defects (fusty, mouldy, muddy, rancid and winey). For this purpose, these defects, available as single standards of the International Olive Council, were added to refined sunflower oil. According to the LDA models and the ANN method, the defected samples were correctly classified. On the other hand, the electronic nose data was used to predict the defect percentage added to sunflower oil using multiple linear regression models. All …
Authentication of extra virgin olive oils by Fourier-transform infrared spectroscopy
2010
Fourier-transform infrared spectroscopy (FTIR), followed by multivariate treatment of the spectral data, was used to classify vegetable oils according to their botanical origin, and also to establish the composition of binary mixtures of extra virgin olive oil (EVOO) with other low cost edible oils. Oil samples corresponding to five different botanical origins (EVOO, sunflower, corn, soybean and hazelnut) were used. The wavelength scale of the FTIR spectra of the oils was divided in 26 regions. The normalized absorbance peak areas within these regions were used as predictors. Classification of the oil samples according to their botanical origin was achieved by linear discriminant analysis (…
TOMOCOMD-CARDD descriptors-based virtual screening of tyrosinase inhibitors: evaluation of different classification model combinations using bond-bas…
2006
Abstract A new set of bond-level molecular descriptors (bond-based linear indices) are used here in QSAR (quantitative structure–activity relationship) studies of tyrosinase inhibitors, for finding functions that discriminate between the tyrosinase inhibitor compounds and inactive ones. A database of 246 compounds was collected for this study; all organic chemicals were reported as tyrosinase inhibitors; they had great structural diversity. This dataset can be considered as a helpful tool, not only for theoretical chemists but also for other researchers in this area. The set used as inactive has 412 drugs with other clinical uses. Twelve LDA-based QSAR models were obtained, the first six us…
A topological substructural approach for the prediction of P-glycoprotein substrates
2006
A topological substructural molecular design approach (TOPS-MODE) has been used to predict whether a given compound is a P-glycoprotein (P-gp) substrate or not. A linear discriminant model was developed to classify a data set of 163 compounds as substrates or nonsubstrates (91 substrates and 72 nonsubstrates). The final model fit the data with sensitivity of 82.42% and specificity of 79.17%, for a final accuracy of 80.98%. The model was validated through the use of an external validation set (40 compounds, 22 substrates and 18 nonsubstrates) with a 77.50% of prediction accuracy; fivefold full cross-validation (removing 40 compounds in each cycle, 80.50% of good prediction) and the predictio…
Chemical Element Levels as a Methodological Tool in Forensic Science
2014
The aim of the present study was to define a methodological strategy for understanding how post- mortem degradation in bones caused by the environment affects different skeletal parts and for selecting better preserved bone samples, employing rare earth elements (REEs) analysis and multivariate statistics. To test our methodological proposal the samples selected belong to adult and young individuals and were obtained from the Late Roman Necropolis of c/Virgen de la Misericordia located in Valencia city centre (Comunidad Valenciana, Spain). Therefore, a method for the determination of major elements, trace elements and REEs in bone remains has been developed employing Inductively-Coupled Pla…
Predicting antitrichomonal activity: A computational screening using atom-based bilinear indices and experimental proofs
2006
Existing Trichomonas vaginalis therapies are out of reach for most trichomoniasis people in developing countries and, where available, they are limited by their toxicity (mainly in pregnant women) and their cost. New antitrichomonal agents are needed to combat emerging metronidazole-resistant trichomoniasis and reduce the side effects associated with currently available drugs. Toward this end, atom-based bilinear indices, a new TOMOCOMD-CARDD molecular descriptor, and linear discriminant analysis (LDA) were used to discover novel, potent, and non-toxic lead trichomonacidal chemicals. Two discriminant functions were obtained with the use of non-stochastic and stochastic atom-type bilinear in…