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

Studies on hybridization effects of different geographic populations of Chlamys farreri - II. The medium-term growth and development of Chlamys farreri populations from China and Russia and their reciprocal crosses

Based on the research of juvenile (2, 3, 4 months) growth and survival of three populations of two different geographic areas in Chlamys farreri from Russian and China and their F, hybrids derived from Chinese cultural population (CC) female x Russian population (RW) male, Chinese wild population (CW) female x Russian population (RW) male, Russian population (RW) female x Chinese wild population (CW) the study of the medium-term (6, 8, 10, 12 months) growth and development of Chlamys farreri was carried out. The four determined results indicated that there existed different extent heterosis (3% similar to 52 %) for the growth in three types of F-1 hybrids, and the offspring derived from CC female X Rmale had a stronger heterosis among the crosses at the medium-term; the uptrend among traits are wet weight > shell width > shell length > shell height, Chinese cultural population could be recognized as excellent parent, and seasonal variations influence very much on the daily increment and growth rate of each trait of Chlamys farreri and it is only able to survive and could barely grow in winter (6 similar to 8 months), but grows fast in temperate season (10similar to12 months).

Geographic variation in the leaf essential oils of Juniperus sabina L. and J-sablina var. arenaria (E.H. Wilson) Farjon

The composition of the leaf oils from seven populations of J. sabina L., one population of Juniperus sabina var. arenaria (E. H. Wilson) Farjon were examined for their geographic variation. In addition, the leaf oils of J. chinensis L. and J. davurica Pall. were compared to J. sabina. Juniperus sabina var. arenarla, the sand loving juniper, oil was found to be very similar to that of J. davurica, Mongolia, and J. sabina, on sand dunes in Mongolia. This suggests that J. sabina var. arenaria might be conspecific with J. davurica. Farjon's move (2001) of J. sabina var. arenaria out of J. chinensis is supported. Considerable differentiation was found in populations of J. sabina from the Iberian peninsula. Cedrol, citronellol, safrole, trans-sabinyl acetate, terpinen-4-ol and beta-thujone were found to be polymorphic in several populations.

复杂裂隙岩体统计断裂本构关系与变形参数研究

Rock mass is widely recognized as a kind of geologic body which consists of rock blocks and discontinuities. The deformation and failure of rock mass is not only determined by rock block，but also by discontinuity which is virtually more important. Mutual cutting and combination of discontinuities controlled mechanical property of rock mass. The complex cutting of discontinuities determine the intense anisotropy on mechanical property of rock mass，especially under the effect of ground stress. Engineering practice has show that the brittle failure of hard rock always occurs when its working stress is far lower than the yield strength and compressive strength，the failure always directly related to the fracture propagation of discontinuities. Fracture propagation of discontinuities is the virtue of hard rock’s failure. We can research the rock mass discontinuous mechanical properties precisely by the methods of statistical analysis of discontinuities and Fracture Mechanics. According to Superposition Principle in Fracture Mechanics，A Problem or C Problem could be chosen to research. Problem A mainly calculates the crack-tip stress field and displacement field on internal discontinuities by numerical method. Problem C calculate the crack-tip stress field and displacement field under the assumption of that the mainly rock mass stress field has been known. So the Problem C avoid the complex mutual interference of stress fields of discontinuities，which is called crack system problem in Fracture Mechanics. To solve Problem C, field test on stress field in the rock mass is needed. The linear Superposition of discontinuities strain energies are Scientific and Rational. The difference of Fracture Mechanics between rock mass and other materials can mostly expression as：other materials Fracture Mechanics mostly face the problem A，and can’t avoid multi-crack puzzle, while the Rock mass Fracture Mechanics answer to the Problem C. Problem C can avoid multi-discontinuities mutual interference puzzle via the ground stress test. On the basis of Problem C, Fracture Mechanics could be used conveniently in rock mass. The rock mass statistics fracture constitutive relations, which introduced in this article, are based on the Problem C and the Discontinuity Strain Energy linear superposition. This constitutive relation has several merits: first, it is physical constitutive relation rather than empirical; second, it is very fit to describe the rock mass anisotropy properties; third, it elaborates the exogenous factors such as ground stress. The rock mass statistics fracture constitutive relation is the available approach to answer to the physical, anisotropic and ground stress impacted rock mass problems. This article stand on the foundation of predecessor’s statistics fractures constitutive relation, and improved the discontinuity distributive function. This article had derived the limitation of negative exponential distribution in the course of regression analysis, and advocated to using the two parameter negative exponential distribution for instead. In order to solve the problems of two-dimension stability on engineering key cross-sectional view in rock mass, this article derived the rock mass planar flexibility tensor, and established rock mass two-dimension penetrate statistics fracture constitutive relation on the basis of penetrate fracture mechanics. Based on the crack tip plasticity research production of penetrate fracture, for example the Irwin plasticity equifinality crack, this article established the way to deal with the discontinuity stress singularity and plastic yielding problem at discontinuity tip. The research on deformation parameters is always the high light region of rock mass mechanics field. After the dam foundation excavation of XiaoWan hydroelectric power station, dam foundation rock mass upgrowthed a great deal of unload cracks, rock mass mechanical property gotten intricacy and strong anisotropy. The dam foundation rock mass mostly upgrowthed three group discontinuities： the decantation discontinuity, the steep pitch discontinuity, and the schistosity plane. Most of the discontinuities have got partial unload looseness. In accordance with ground stress field data, the dam foundation stress field greatly non-uniform, which felled under the great impaction of tectonic stress field, self-weight stress field, excavation geometric boundary condition, and excavation, unload. The discontinuity complexity and stress field heterogeneity, created the rock mass mechanical property of dam foundation intricacy and levity. The research on the rock mass mechanics, if not take every respected influencing factor into consideration as best as we can, major errors likely to be created. This article calculated the rock mass elastic modulus that after Xiao Wan hydroelectric power station dam foundation gutter excavation finished. The calculation region covered possession monolith of Xiao Wan concrete double-curvature arch dam. Different monolith were adopted the penetrate fracture statistics constitutive relation or bury fracture statistics constitutive relation selectively. Statistics fracture constitutive relation is fit for the intensity anisotropy and heterogeneity rock mass of Xiao Wan hydroelectric power station dam foundation. This article had contrastive analysis the statistics fracture constitutive relation result with the inclined plane load test actual measurement elastic modulus and RMR method estimated elastic modulus, and find that the three methods elastic modulus have got greatly comparability. So, the statistics fracture constitutive relations are qualified for trust. Generally speaking，this article had finished following works based on predecessors job: “Argumentation the C Problems of superposition principle in Fracture Mechanics, establish two-dimension penetrate statistics fracture constitutive relation of rock mass, argue the negative exponential distribution limitation and improve it, improve of the three-dimension berry statistics fracture constitutive relation of rock mass, discontinuity-tip plastic zone isoeffect calculation, calculate the rock mass elastic modulus on two-dimension cross-sectional view”. The whole research clue of this article inherited from the “statistics rock mass mechanics” of Wu Faquan（1992）.

The Discontinuity of Cultural Process in the Digital Civilization. Main Dangers

Keynote presentation on ETHICOMP2001.

Analiza ilościowa degradacji szlaków turystycznych na przykładzie Gorczańskiego Parku Narodowego i Popradzkiego Parku Krajobrazowego

Wydział Nauk Geograficznych i Geologicznych

Self perceived health status of schizophrenic patients in Spain: an analysis of geog raphical differences using bayesian approach

Objectives. This paper explores the use of regression models for estimating health status of schizophrenic patients, from a Bayesian perspective. Our aims are: 1- To obtain a set of values of health states of the EQ-5D based on self-assessed health from a sample of schizophrenic patients. 2- To analyse the differences in the health status and in patients’ perceptions of their health status between four mental-health districts in Spain. Methods. We develop two linear models with dummy variables. The first model seeks to obtain an index of the health status of the patients using a VAS as a dependent variable and the different dimensions of EQ-5D as regressors. The second model allows to analyse the differences between the self-assessed health status in the different geographic areas and also the differences between the patients’ self-assessed health states, irrespective of their actual health state, in the different geographic areas. The analysis is done using Bayesian approach with Gibbs sampling (computer program WinBUGS 1.4). Data concerning self-assessed EQ-5D with VAS from four geographic areas of schizophrenic patients were obtained for the purposes of this analysis. Results. We obtained the health status index for this sample and analysed the differences for this index between the four geographic areas. Our study reveals variables that explain the differences in patients’ health status and differences in their health states assessment. We consider four possible scenarios.

On the Geographic Location of Internet Resources

One relatively unexplored question about the Internet's physical structure concerns the geographical location of its components: routers, links and autonomous systems (ASes). We study this question using two large inventories of Internet routers and links, collected by different methods and about two years apart. We first map each router to its geographical location using two different state-of-the-art tools. We then study the relationship between router location and population density; between geographic distance and link density; and between the size and geographic extent of ASes. Our findings are consistent across the two datasets and both mapping methods. First, as expected, router density per person varies widely over different economic regions; however, in economically homogeneous regions, router density shows a strong superlinear relationship to population density. Second, the probability that two routers are directly connected is strongly dependent on distance; our data is consistent with a model in which a majority (up to 75-95%) of link formation is based on geographical distance (as in the Waxman topology generation method). Finally, we find that ASes show high variability in geographic size, which is correlated with other measures of AS size (degree and number of interfaces). Among small to medium ASes, ASes show wide variability in their geographic dispersal; however, all ASes exceeding a certain threshold in size are maximally dispersed geographically. These findings have many implications for the next generation of topology generators, which we envisage as producing router-level graphs annotated with attributes such as link latencies, AS identifiers and geographical locations.