Search

Organoleptic evaluations in a 10-point scale of wine experts and experimental physicochemical parameters of red (1599 samples) and white (4898 samples) wines of Portuguese manufacturers were analyzed using STATISTICA sofware. Methods of agglomerative and iterative (k-means algorithm) clustering revealed the grouping of similar wine samples into three, four and six clusters depending on the Euclidean distance of association. The quantitative filling of clusters with samples of bad wines (grades 3 and 4), normal wines (grades 5 and 6) and good wines (grades 7, 8, 9) was established. Neural network (MLP) and discriminant analyzes (DA) were performed; algorithms of classification trees (CT), support vector machines (SVM), naive Bayesian classification (NBC) and nearest neighbors (kNN) were involved. The best performance was demonstrated by neural network models. Multilayer perceptorons classifiers were trained: for red wines—MLP 11-7-3, MLP 11-13-4, MLP 11-14-6; for white wines—MLP 11-9-3, MLP 11-5-4, MLP 11-9-6. The properties of wines, whose contribution to the separating power of classifiers is decisive, are revealed. The ranges of changes in physicochemical parameters in three clusters of red and white wines for bad, normal and good wines were given. [ABSTRACT FROM AUTHOR]

The data on 178 samples of Italian red wines taken from the public machine learning repository UCI have been studied. Computer analysis of 13 physico-chemical properties of the wine samples on their distribution between three groups has been performed via different data mining methods. Classification models: factor, discriminant, canonical, and neural network (multilayer perceptron MLP, Kohonen's map SOFM) ones, predicting models (support vector machine, Bayesian classifier, and nearest neighbor method), and decision trees have been built. The neural network classifiers SOFM 13-3, MLP 13-5-3, and SOFM 16-3 have been trained. It has been found that predicting power of the models is determined by the following variables: proline, flavonoids, color intensity, proteins, and alcohol. [ABSTRACT FROM AUTHOR]

Computer analysis of the thermodynamic constants of complexation of [2.2.2]-cryptand with alkali metal cations (M[222]+ cryptates with M = Li, Na, K, Rb, and Cs) in water and organic solvents such as methanol, ethanol, 1-propanol, acetonitrile, benzonitrile, acetone, N,N-dimethylformamide, N-methylpyrrolidone, nitrobenzene, nitromethane, 1,2-dichloroethane, and propylene carbonate at 298.15 K has been performed. Exploratory (factorial, cluster, discriminant, canonical, decision tree), regression and neural network models of effects of the properties of solvents and cations on the cation cryptates stability have been built. The neural network approximator MLP 4-7-1 and the classifiers of the stability constants of cryptates – the multilayer perceptron MLP 4-7-4 and the self-organizing Kohonen map SOFM 8-4 – have been trained. Independent data on the stability constants of alkali metal cations cryptates demonstrate the predictive capabilities of the trained MLP 4-7-1 perceptron approximator. [ABSTRACT FROM AUTHOR]

Exploratory, regression, and neural network analysis of the stability constants of crown ether [12C4, 16C5, (CH3)216C5, DB21C7, DB24C8, DCH24C8, DB30C10] 1 : 1 complexes with alkaline (Li+, Na+, K+, Cs+, Rb+), alkaline-earth (Ca2+, Sr2+, Ba2+), and heavy (Ag+, Tl+, Co2+, Cu2+, Pb2+) metals and NH4+ in water and organic solvents (methanol, acetonitrile, acetone, N,N-dimethylformamide, nitrobenzene, nitromethane, 1,2-dichloroethane, propylene carbonate) at 298.15 K obtained via conductometry has been performed. Factor, cluster, discriminant, canonical, decision tree, regression, and neural network models of clustering, approximation, and prediction of thermodynamic constants of the complexation depending on the properties of the ligand, the cation, and the solvent have been developed. The trained MLP 7-5-5 Multilayer Perceptron Cluster has completely confirmed the k-means clustering. Independent data on the stability constants of coronates have demonstrated the predictive capacity of the trained perceptron-approximator MLP 7-7-1. [ABSTRACT FROM AUTHOR]

Exploration (factor, cluster, and decision tree), regression (multiple linear regression), and neural network (regression, classification) models of clustering, approximation and prediction of the stability constants of cation complexes with ionophore antibiotics (nonactin, monactin, dinactin, trinactin, ennatin B, monensin A, and valinomycin) according to the properties of organic solvents (methanol, ethanol, acetonitrile, and nitrobenzene) and cations (Li+, Na+, K+, Rb+, Cs+, Tl+, Ag+, NH4+, Mg2+, Ca2+, Sr2+, Ba2+, and Mn2+) have been developed. It has been shown that neural network performance is better than that of multiple linear regression (the correlation coefficient on the training sample 0.756 compared to 0.697). The neural network classifier and the classification tree has confirmed the clustering of stability of ionophore–cation complexes carried out by the exploratory k-means method by 97.2%. The prognostic capabilities of the constructed multilayer perceptron have been demonstrated. [ABSTRACT FROM AUTHOR]

An approach to predict the stability constants of coronates from the properties of solvents, cations, and crown ethers has been developed based on exploratory and neural network methods for mathematical modeling of equilibria in solutions. Exploratory (factor, cluster, discriminant, canonical, decision trees), regression, and neural network (supervised and Kohonen network) models of the stability of crown ethers (12С4, 13С4, 14С4, 15С4, 15С5, 18С6, 21С7, 24С8, B12C4, B15C5, CH15C5, CH18C6, DCH18C6, DCH21C7, DB18C6, DB21C7, DB24C8, DB27C9, and DB30C10) complexes with cations of alkali (Li+, Na+, K+, Cs+, Rb+) and alkaline-earth (Ca2+, Sr2+, Ba2+) metals in aqueous and non-aqueous (acetone, acetonitrile, dimethyl sulfoxide, methanol, pyridine, dimethylformamide, dioxane, propylene carbonate, 1,2-dichloroethane, and nitrobenzene) solutions have been developed according to the properties of solvents (diameter of solvent molecule, Kamlet–Taft parameter, Dimroth–Reichardt parameter, dielectric constant), crown ethers (Balaban topological index), and cations (cation diameter) at 298.15 K. [ABSTRACT FROM AUTHOR]

Models of multiple linear regression and multilayer artificial neural network have been developed for modeling and predicting the stability constants of sodium and potassium coronates basing on the properties of aqueous-organic solvents (water-methanol, water-propan-2-ol, water-acetonitrile, and water-acetone). The values of the coronates stability constants in water-ethanol solvents have been predicted, and the predictions of the models of multiple linear regression and an artificial neural network models have been compared. The contributions of electrostatic, cohesive, and electron-donating interactions to the increase in the stability of the coronates have been quantitatively assessed basing on the models of multiple linear regression and the principle of free energies linearity. Neural network models based on unsupervised (multilayer perceptrons) and supervised (Kohonen networks) learning algorithms have been developed to classify the stability of sodium and potassium coronates. The neural network classifiers have fully confirmed the classification of the coronated stability via the k-means exploration method. [ABSTRACT FROM AUTHOR]

Linear Fisher classification functions, canonical linear discriminant functions, and trees (rules) for classification and prediction of the stability of sodium (18-crown-6Na+) and potassium coronates (18-crown-6K+) from the properties of aqueous-organic solvents (water—methanol, water—propan-2-ol, and water—acetonitrile) have been obtained basing on multivariate exploratory analysis. The proposed approach to prediction of the coronates stability has been verified using independent experimental data on the stability constants of sodium and potassium coronates in water—dioxane and water—acetone mixtures. The constructed classification functions and rules have possessed high predictive potential. [ABSTRACT FROM AUTHOR]

The effect of dielectric and chemical (donor-acceptor interactions, cohesion energy density) properties of aqueous-organic solvents (water-methanol, water-2-propanol, water-acetonitrile, water-acetone, water-dioxane) on the thermodynamic functions of complexation of sodium and potassium cations with 18-crown-6 ether at 298 K was subjected to overall correlation analysis. [ABSTRACT FROM AUTHOR]