Statistical Linear Discriminant Analysis Research Articles

Effective Descriptor for Screening Single-Molecule Conductance Switches.

The adsorption-type molecular switch exhibits bistable states with an equivalently long lifetime at the organic/inorganic interface, promising reliable switching behavior and superior assembly ability in the electronic circuits at the molecular scale. However, the number of reported adsorption-type molecular switches is currently less than 10, and exploring these molecular switches poses a formidable challenge due to the intricate interplay occurring at the interface. To address this challenge, we have developed a model enabling the identification of diverse molecular switches on metal surfaces based on easily accessible physical characteristics. These characteristics primarily include the metal valency electron concentration, the work function of metal surfaces, and the electronegativity difference of molecules. Using this model, we identified 56 new molecular switches. Employing the gradient descent algorithm and statistical linear discriminant analysis, we constructed an explicit descriptor that establishes a relationship between the interfacial structure and chemical environment and the stability of molecular switches. The model's accuracy was validated through density functional theory calculations, achieving a 90% accuracy for aromatic molecular switches. The conductive switching behaviors were further confirmed by nonequilibrium Green's function transport calculations.

Molecular Topology QSAR Strategy for Crop Protection: New Natural Fungicides with Chitin Inhibitory Activity

Nowadays, crop protection is a major concern and how to proceed is a delicate point of contention. New products must be safe and ecofriendly in accordance with the actual legislation. In this context, we present a quantitative structure–activity relationship strategy based on molecular topology as a tool for generating natural products as potential fungicides following a mechanism of action based on the synthesis of chitin inhibition (chitinase inhibition). Two discriminant equations using statistical linear discriminant analysis were used to identify three potential candidates (1-methylxanthine, hematommic acid, and antheraxanthin). The equations showed accuracy and specificity levels above 80%, minimizing the risk of selecting false active compounds.

Detection ofFusarium oxysporumFungal Isolates Using ATR Spectroscopy

Fungi are considered as serious pathogens for many plants, potentially causing severe economic damage. Early detection and identification of these pathogens is crucial for their timely control. The methods available for identification of fungi are time consuming and not always very specific. In this study, the potential of FTIR-ATR spectroscopy was examined together with advanced mathematical principle component analysis (PCA) and statistical linear discriminant analysis (LDA) to differentiate among 10 isolates of Fusarium oxysporum. The results are encouraging and indicate that FTIR-ATR can successfully detect different isolates of Fusarium oxysporum. Based on PCA and LDA calculations in the region 850–1775 cm-1with 16 PC's, the different strains from the same fungal genus could be classified with 75.3% and 69.5% success rates using the “leave one out” method and “20–80% algorithm” respectively.

Vitamin C and Sugar Levels as Simple Markers for Discriminating Spanish Honey Sources

In this work, 7 Spanish honeys with different botanical origins were studied. The honey origins were rosemary, chestnut, lavender, echium, thyme, multifloral, and honeydew. The chemical compounds determined were ascorbic acid (vitamin C), hydroxymethylfurfural, and major sugar contents (glucose and fructose). The physicochemical parameters, pH, conductivity, moisture, free acidity, and color, were also measured. Vitamin C is an important antioxidant in food, and the possibility to use it as discriminate parameter among different honeys was studied. The determination of vitamin C in honey samples was carried out by 2 different methods, volumetric and chromatographic comparing the results by both statistically. Vitamin C content was higher in thyme honeys than in the other types; however a wide dispersion in the values was found. Through a linear discriminant analysis (LDA), conductivity, glucose, fructose, and vitamin C content were the most important discriminant parameters. Vitamin C content in different honey sources has been determined by a simple and rapid chromatographic method (less than 3 min) in honeys from 6 botanical origins. The results together with glucose and fructose content and some physicochemical parameters have been studied in order to discriminate the botanical origin of honeys and in the future certified their quality. A statistical LDA was applied to the data, and differentiation of honey sources was possible with very good agreement. The vitamin C content found in thymus honeys was significantly higher than in other types. This fact makes vitamin C a special marker for thymus honeys that have a higher antioxidant effect than the others giving it special properties. The identification of honey sources is essential for beekeepers in order to certify honeys for consumers.

A misclassification cost‐minimizing evolutionary–neural classification approach

AbstractMachine learning algorithms that incorporate misclassification costs have recently received considerable attention. In this paper, we use the principles of evolution to develop and test an evolutionary/genetic algorithm (GA)‐based neural approach that incorporates asymmetric Type I and Type II error costs. Using simulated, real‐world medical and financial data sets, we compare the results of the proposed approach with other statistical, mathematical, and machine learning approaches, which include statistical linear discriminant analysis, back‐propagation artificial neural network, integrated cost preference‐based linear mathematical programming‐based minimize squared deviations, linear integrated cost preference‐based GA, decision trees (C 5.0, and CART), and inexpensive classification with expensive tests algorithm. Our results indicate that the proposed approach incorporating asymmetric error costs results in equal or lower holdout sample misclassification cost when compared with the other statistical, mathematical, and machine learning misclassification cost‐minimizing approaches. © 2006 Wiley Periodicals, Inc. Naval Research Logistics, 2006.

2D-LDA: A statistical linear discriminant analysis for image matrix

This paper proposes an innovative algorithm named 2D-LDA, which directly extracts the proper features from image matrices based on Fisher’s Linear Discriminant Analysis. We experimentally compare 2D-LDA to other feature extraction methods, such as 2D-PCA, Eigenfaces and Fisherfaces. And 2D-LDA achieves the best performance.

A threshold-varying artificial neural network approach for classification and its application to bankruptcy prediction problem

We propose a threshold-varying artificial neural network (TV-ANN) approach for solving the binary classification problem. Using a set of simulated and real-world data set for bankruptcy prediction, we illustrate that the proposed TV-ANN fares well, both for training and holdout samples, when compared to the traditional backpropagation artificial neural network (ANN) and the statistical linear discriminant analysis. The performance comparisons of TV-ANN with a genetic algorithm-based ANN and a classification tree approach C4.5 resulted in mixed results.

Linear models for minimizing misclassification costs in bankruptcy prediction

AbstractThis paper illustrates how a misclassification cost matrix can be incorporated into an evolutionary classification system for bankruptcy prediction. Most classification systems for predicting bankruptcy have attempted to minimize misclassifications. The minimizing misclassification approach assumes that Type I and Type II error costs for misclassifications are equal. There is evidence that these costs are not equal and incorporating costs into the classification systems can lead to better and more desirable results. In this paper, we use the principles of evolution to develop and test a genetic algorithm (GA) based approach that incorporates the asymmetric Type I and Type II error costs. Using simulated and real‐life bankruptcy data, we compare the results of our proposed approach with three linear approaches: statistical linear discriminant analysis (LDA), a goal programming approach, and a GA‐based classification approach that does not incorporate the asymmetric misclassification costs. Our results indicate that the proposed approach, incorporating Type I and Type II error costs, results in lower misclassification costs when compared to LDA and GA approaches that do not incorporate misclassification costs. Copyright © 2001 John Wiley & Sons, Ltd.

Application of Bayesian Network Classifiers and Data Envelopment Analysis for Mining Breast Cancer Patterns

Data envelopment analysis (DEA) and learning Bayesian networks (LBN) are relatively unexplored approaches for data mining. Recent theoretical research indicates that DEA and LBN can be used for certain data mining applications. We apply DEA and LBN for discovering the breast cancer pattern. We use data from a large hospital in Pennsylvania to discover the breast cancer patterns, and benchmark the performance of DEA and LBN against a popular statistical linear discriminant analysis technique. The results of our experiments indicate that DEA and LBN outperform statistical linear discriminant analysis.

Spatio-temporal feature maps using gated neuronal architecture

In this paper, Kohonen's self-organizing feature map is modified by a novel technique of allowing the neurons in the feature map to compete in a selective manner. The selective competition is achieved by grating the N-dimensional feature space using a spatial frequency and setting a criterion for the neurons to compete based on the region in which the input pattern resides. The spatial grating and selective competition are achieved by introducing a gated neuronal architecture in the feature map. As the selection criterion changes with time, it generates a time sequence of winning node indexes providing more input information and potentially allowing higher classification performance. These time sequences are then used to predict the class label of the input pattern more accurately. Three possible class label prediction algorithms are formulated based on evidential reasoning method and Bayes conditional probability theorem. These are tested on real world 8-class texture and a synthetic 12-class 3D object recognition problems. The classification performance is then compared with the results obtained by using a standard statistical linear discriminant analysis.

A linear goal programming model for classification with non-monotone attributes

Classification function analysis concerns separating two or more groups of objects in a data set and allocating new objects to previously defined groups. Usually, a set of attribute weights are estimated and the classification decision of an object is based on the weighted sum of its attribute scores. Statistical linear discriminant analysis, logistic regression, and linear programming approaches to classification problems have been proposed to address this problem. However, monotonicity of the attribute scores with respect to the likelihood of belonging to one specific group is presumed by these approaches. This may not be realistic in many applications. In this paper, a linear goal programming approach with the ability to capture the non-monotonicity of some attribute scores in classification problems is proposed. Classification performances of this approach and other classification approaches are evaluated by a simulation experiment. The results are very encouraging for the proposed approach.

Enological Parameters and Pattern Recognition Methods in the Geographic Differentiation of Spanish Red Wines

Two hundred thirty wines, 1976 to 1983 vintages, from four different Spanish regions have been analyzed by pattern recognition methods in order to characterize the groups and classify unknown samples according to their geographic origin. The nine physico-chemical parameters determined are termed inexpensive variables since they can be evaluated at any enological station. The variables lactic, tartaric, and malic acids, titratable acidity, and potassium content are the most relevant in this study. The effect of factors such as varietal diversity, climatic conditions, or winemaking techniques, which account for the differences in the mean values of the variables for each region, are discussed. Coincident classification results have been found for all three supervised methods of analysis used: statistical linear discriminant analysis (SLDA), linear learning machine (LLM), and K-nearest neighbor (KNN). The correct classification percentage obtained has been <i>ca</i> 90% for all procedures employed.

Potential methods in pattern recognition: Part 4. A combination of ALLOC and Statistical Linear Discriminant Analysis

It is shown that ALLOC when associated with descriptive statistical linear discriminant analysis (“display” SLDA) is in some situations a better alternative than ALLOC and SLDA for classification purposes (“classification” SLDA). It is possible with a combination of “display” SLDA and ALLOC to adjust the suboptimal decision boundaries of “classification” SLDA when the underlying assumptions about the data in “classification” SLDA are not fulfilled, whereas the combination of “display” SLDA with ALLOC permits classification of new objects without computer assistance which is not possible with ALLOC alone. The performance of “display” SLDA—ALLOC is discussed on the basis of the differentiations between THYROID functional states.

Analytical Parameters in the Characterization of Three Venetian Wines. Application of the Statistical Linear Discriminant Analysis

The possibilities of discriminating between the Venetian D.O.C. wines Soave Classico (S), Prosecco di Conegliano e Valdobbiadene (P), and Verduzzo del Piave (V) are tested on the basis of the following analysis determinations: sodium, potassium, calcium, magnesium, chloride, pH, total acidity, phosphorus, ash content, and alkalinity of the ashes. Comparisons are drawn by the statistical method of linear discriminant analysis. Optimal discriminant functions and the threshold values are reported for the comparisons of S-P, S-V, and P-V; the probabilities of classification errors are also estimated. For the comparisons of S-P and S-V a considerable discriminating power (error 1.4% and 3.7%, respectively) is obtained by combining the parameters pH, ash content, and alkalinity of the ashes. The comparison of P-V is tentatively achieved by combining the parameters sodium, potassium, and chloride; however, the result has shown that the discriminating test is not satisfactory.

The application of linear discriminant analysis in the diagnosis of thyroid diseases

The effectiveness of five in-vitro laboratory tests for differentiation between three thyroid functional states (EU, HYPO and HYPER thyroidism) has been determined by using statistical linear discriminant analysis. The optimal linear combination of laboratory tests obtained by means of linear discriminant analysis results in a better use of the information present in each test, so that the possible redundancy of tests can be assessed. In this context, some feature selection criteria were evaluated. It is shown that in this application only two laboratory tests are necessary to obtain a sufficiently high diagnostic effectiveness when linear discriminant analysis is applied.