Semisupervised Dimensionality Reduction and Classification Through Virtual Label Regression

Semisupervised dimensionality reduction has been attracting much attention as it not only utilizes both labeled and unlabeled data simultaneously, but also works well in the situation of out-of-sample. This paper proposes an effective approach of semisupervised dimensionality reduction through label propagation and label regression. Different from previous efforts, the new approach propagates the label information from labeled to unlabeled data with a well-designed mechanism of random walks, in which outliers are effectively detected and the obtained virtual labels of unlabeled data can be well encoded in a weighted regression model. These virtual labels are thereafter regressed with a linear model to calculate the projection matrix for dimensionality reduction. By this means, when the manifold or the clustering assumption of data is satisfied, the labels of labeled data can be correctly propagated to the unlabeled data; and thus, the proposed approach utilizes the labeled and the unlabeled data more effectively than previous work. Experimental results are carried out upon several databases, and the advantage of the new approach is well demonstrated.

Selection for variables in quantitative structure-activity/property relationship studies - Comparison between methods of orthogonal descriptors and leaps-and-bounds regression analysis

In this paper, the comparison of orthogonal descriptors and Leaps-and-Bounds regression analysis is performed. The results obtained by using orthogonal descriptors are better than that obtained by using Leaps-and-Bounds regression for the data set of nitrobenzenes used in this study. Leaps-and-Bounds regression can be used effectively for selection of variables in quantitative structure-activity/property relationship(QSAR/QSPR) studies. Consequently, orthogonalisation of descriptors is also a good method for variable selection for studies on QSAR/QSPR.

Studies on the method of leaps and bounds regression

In this paper, we introduce the method of leaps and bounds regression which can be used to select variables quickly and obtain the best regression models. These models contain one variable, two variables, three variables and so on. The results obtained by using leaps and bounds regression were compared with those achieved by using stepwise regression to lead to the conclusion that leaps and bounds regression is an effective method.

Prediction of molar absorptivities of color reagents during their color reactions with cerium using multiple regression analysis and neural network

A quantitative structure-property study has been made on the relationship between molar absorptivities (epsilon) of asymmetrical phosphone bisazo derivatives of chromotropic acid and their color reactions with cerium by multiple regression analysis and neural network. The new topological indices A(x1) - A(x3) suggested in our laboratory and molecular connectivity indices of 43 compounds have been calculated. The results obtained from the two methods are compared. The neural network model is superior to the regression analysis technique and gave a prediction which was sufficiently accurate to estimate the molar absorptivities of color reagents during their color reactions with cerium.

Study on Phenol Compounds Using Multiple Regression and Neural Network Methods

In this paper, the molecular connectivity indices and the electronic charge parameters of forty-eight phenol compounds nave been calculated. and applied for studying the relationship between partition coefficients and structure of phenol compounds. The results demonstrate that the properties of compounds can be described better with selective parameters, and the results obtained by neural network are superior to that by multiplle regression.

Quantitative structure-property relationships for colour reagents and their colour reactions with ytterbium using regression analysis and computational neural networks

In this paper, the new topological indices A(x1)-A(x3) suggested in our laboratory and molecular connectivity indices have been applied to multivariate analysis in structure-property studies. The topological indices of twenty asymmetrical phosphono bisazo derivatives of chromotropic acid have been calculated. The structure-property relationships between colour reagents and their colour reactions with ytterbium have been studied by A(x1)-A(x3) indices and molecular connectivity indices with satisfactory results. Multiple regression analysis and neural networks were employed simultaneously in this study.

QUANTITATIVE STRUCTURE-ACTIVITY-RELATIONSHIPS FOR TOXICITY OF PHENOLS USING REGRESSION-ANALYSIS AND COMPUTATIONAL NEURAL NETWORKS

Quantitative structure-toxicity models were developed that directly link the molecular structures of a et of 50 alkYlated and/or halogenated phenols with their polar narcosis toxicity, expressed as the negative logarithm of the IGC50 (50% growth inhibitor

Altitudinal differences in the leaf fitness of juvenile and mature alpine spruce trees (Picea crassifolia)

In many plant species, leaf morphology varies with altitude, an effect that has been attributed to temperature. It remains uncertain whether such a trend applies equally to juvenile and mature trees across altitudinal gradients in semi-arid mountain regions. We examined altitude-related differences in a variety of needle characteristics of juvenile (2-m tall) and mature (5-m tall) alpine spruce (Picea crassifolia Kom.) trees growing at altitudes between 2501 and 3450 m in the Qilian Mountains of northwest China. We found that stable carbon isotope composition (delta C-13), area- and mass-based leaf nitrogen concentration (N-a, N-m), number of stomata per gram of nitrogen (St/N), number of stomata per unit leaf mass (St/LM), projected leaf area per 100 needles (LA) and leaf mass per unit area (LMA) varied nonlinearly with altitude for both juvenile and mature trees, with a relationship reversal point at about 3 100 m. Stomatal density (SD) of juvenile trees remained unchanged with altitude, whereas SD and stomatal number per unit length (SNL) of mature spruce initially increased with altitude, but subsequently decreased. Although several measured indices were generally found to be higher in mature trees than in juvenile trees, N-m, leaf carbon concentration (C.), leaf water concentration. (LWC), St/N, LA and St/LM showed inconsistent differences between trees of different ages along the altitudinal gradient. In both juvenile and mature trees, VC correlated significantly with LMA, N-m, N-a, SNL, St/LM and St/N. Stomatal density, LWC and LA were only significantly correlated with delta C-13 in mature trees. These findings suggest that there are distinct ecophysiological differences between the needles of juvenile and mature trees that determine their response to changes in altitude in semi-arid mountainous regions. Variations in the fitness of forests of different ages may have important implications for modeling forest responses to changes in environmental conditions, such as predicted future temperature increases in high attitude areas associated with climate change.

A Constant-Factor Approximation Algorithm for Embedding Unweighted Graphs into Trees

We present a constant-factor approximation algorithm for computing an embedding of the shortest path metric of an unweighted graph into a tree, that minimizes the multiplicative distortion.

Hierarchical Multi-classification with Predictive Clustering Trees in Functional Genomics

Struyf, J., Dzeroski, S. Blockeel, H. and Clare, A. (2005) Hierarchical Multi-classification with Predictive Clustering Trees in Functional Genomics. In proceedings of the EPIA 2005 CMB Workshop

Fuzzy-Rough Feature Significance for Fuzzy Decision Trees

R. Jensen and Q. Shen, 'Fuzzy-Rough Feature Significance for Fuzzy Decision Trees,' in Proceedings of the 2005 UK Workshop on Computational Intelligence, pp. 89-96, 2005.

Index Trees for Efficient Deformable Shape-based Retrieval

An improved method for deformable shape-based image indexing and retrieval is described. A pre-computed index tree is used to improve the speed of our previously reported on-line model fitting method; simple shape features are used as keys in a pre-generated index tree of model instances. In addition, a coarse to fine indexing scheme is used at different levels of the tree to further improve speed while maintaining matching accuracy. Experimental results show that the speedup is significant, while accuracy of shape-based indexing is maintained. A method for shape population-based retrieval is also described. The method allows query formulation based on the population distributions of shapes in each image. Results of population-based image queries for a database of blood cell micrographs are shown.

Leaf galls in our native trees and shrubs

Plant galls constitute a branch of study and research which has been to me a subject of much interest for some time. At the start of this work, it was intended to include Plant galls in general, but after some months this was found to be too comprehensive a field and would in fact take a great many years to study fully. Even leaf galls alone, both of herbs and trees provide so large a field of investigation that ultimately I decided to confine my attention to those or our native trees and shrubs. Upon looking up the literature on this subject, it will be found that in nearly all cases, either the gall is described fully and mere mention made or the agent concerned in its production, or vice versa. This state of things is most unsatisfactory, as in studying galls, both the gall-maker and the gall formation must be examined in detail before it is safe to apply nomenclature. This work, therefore, sets out to give accurate and scientific descriptions of both galls and gall-makers. The difficulties encountered are manifold; firstly, our trees are all deciduous, hence, the collecting period is necessarily restricted to that time of the year between the appearance of the buds and the fall of the leaf. Secondly, the rearing of imagines is always difficult, especially in the case or the autumn gall; more will be said on this matter later. Lastly, due to war-time conditions much trouble was experienced in obtaining suitable literature and many invaluable books on this subject were unprocurable. The Plates at the back have all been copied from original material except in the case or the Phytoptid mites which have been sketched with the help of illustrations, the reason for this being the difficulty of making suitable mounts of these minute creatures, Where possible all stages or at least larva and imago have been sketched, together with the host plant and the type of gall-formation produced. Slides have also been made of most larvae and the imagines attached to cards and pinned on to pith or cork in the usual manner.

Power system parameters forecasting using Hilbert-Huang transform and machine learning

A novel hybrid data-driven approach is developed for forecasting power system parameters with the goal of increasing the efficiency of short-term forecasting studies for non-stationary time-series. The proposed approach is based on mode decomposition and a feature analysis of initial retrospective data using the Hilbert-Huang transform and machine learning algorithms. The random forests and gradient boosting trees learning techniques were examined. The decision tree techniques were used to rank the importance of variables employed in the forecasting models. The Mean Decrease Gini index is employed as an impurity function. The resulting hybrid forecasting models employ the radial basis function neural network and support vector regression. A part from introduction and references the paper is organized as follows. The second section presents the background and the review of several approaches for short-term forecasting of power system parameters. In the third section a hybrid machine learningbased algorithm using Hilbert-Huang transform is developed for short-term forecasting of power system parameters. Fourth section describes the decision tree learning algorithms used for the issue of variables importance. Finally in section six the experimental results in the following electric power problems are presented: active power flow forecasting, electricity price forecasting and for the wind speed and direction forecasting.

Weighted and two-stage least squares estimation of semiparametric truncated regression models

This paper provides a root-n consistent, asymptotically normal weighted least squares estimator of the coefficients in a truncated regression model. The distribution of the errors is unknown and permits general forms of unknown heteroskedasticity. Also provided is an instrumental variables based two-stage least squares estimator for this model, which can be used when some regressors are endogenous, mismeasured, or otherwise correlated with the errors. A simulation study indicates that the new estimators perform well in finite samples. Our limiting distribution theory includes a new asymptotic trimming result addressing the boundary bias in first-stage density estimation without knowledge of the support boundary. © 2007 Cambridge University Press.