Splitting Criterion Research Articles

Simulation of fracture/breakup of concrete magazine using cohesive element

To determine the potential hazard zone around an ammunition storage magazine is the research interest of many researchers working on explosive safety. Carrying out an explosion test of an ammunition storage magazine is expensive. As such, numerical modeling offers an economical solution to supplement the field tests, and it has been widely used to simulate the magazine breakup. In the literature and own previous studies, numerical simulations often resorted to the Lagrangian finite‐element method and used the nodal‐split algorithm to model the concrete fracture/breakup upon internal detonation. Pitfalls in that approach have not been fully recognized [–]. In it, all elements initially connected to the same node would be spilt in all directions once the nodal strain reached the threshold. This strain‐based splitting criterion lacks of sound physical meaning. To overcome those pitfalls, the present study adopts a kind of interface element, namely a cohesive element, to simulate the fracture/breakup. Strain rate dependent experiments were carried out to obtain the fracture criteria, which were then applied to define the threshold for element‐face split. Results show that it yields good agreement with experiment tests, including Split Hopkinson Pressure Bar (SHPB) on concrete specimen and breakup of concrete magazine upon detonation.

Phreatophytic vegetation responses to groundwater depth in a drying mediterranean‐type landscape

AbstractQuestionsDoes the plant hydrotype composition of phreatophytic vegetation alter with drawdown of the water table? Can alternative states of phreatophytic vegetation be identified, and do they persist despite a multi‐decadal drying of a mediterranean‐type landscape?LocationField study, coastal sandy plain, Gnangara Groundwater System, south‐western Australia.MethodsWe used a data set spanning 35 yr to determine alternative states of phreatophytic vegetation and how they relate to spatial variation and temporal change in hydrological habitat. Multivariate regression trees (MRT) were used to identify the range in depth to groundwater associated with vegetation states (characteristic hydrotype compositions). Logistic regression analyses were conducted to predict the proportional representation of each hydrotype. Hydrotype compositions of recent observations were correlated against modelled values derived from historic observations to test whether the relationship between vegetation states and supporting hydrological habitats remains valid after significant decline in the water table.ResultsThe MRT analyses for the historic and recent observations produced a main split criterion at 2.0 and 2.9 m groundwater depth, respectively, essentially dividing records into those dominated by hydrophytes or xerophytes. No hydrophytes were present at groundwater depths >10.5 m (>8.5 m for recent observations), where xerophytes and generalists were dominant. Only extreme groundwater depths were dominated by a single hydrotype class (hydrophytes or xerophytes), while intermediate clusters represented a transition across the hydrotype classes. The logistic regressions of the proportion of hydrotypes as a function of groundwater depth resulted in good fits for hydrophytes and xerophytes only.ConclusionsVegetation states were consistent over 35 yr of progressive increase in depth to water table. Distribution of alternative states along a gradient in groundwater depths was exponential and was reflected in the distribution of hydrophytes and mesophytes. The ranges of groundwater depths at which individual states occur suggest that as groundwater depth increases, the degree of habitat specificity decreases. The narrow range of groundwater depths within which hydrophytes and mesophytes are dominant suggests that small changes in depth could lead to transitions between states. With ongoing drying of the landscape, persistence of these vegetation states can be expected to diminish.

OCCT: A One-Class Clustering Tree for Implementing One-to-Many Data Linkage

One-to-many data linkage is an essential task in many domains, yet only a handful of prior publications have addressed this issue. Furthermore, while traditionally data linkage is performed among entities of the same type, it is extremely necessary to develop linkage techniques that link between matching entities of different types as well. In this paper, we propose a new one-to-many data linkage method that links between entities of different natures. The proposed method is based on a one-class clustering tree (OCCT) that characterizes the entities that should be linked together. The tree is built such that it is easy to understand and transform into association rules, i.e., the inner nodes consist only of features describing the first set of entities, while the leaves of the tree represent features of their matching entities from the second data set. We propose four splitting criteria and two different pruning methods which can be used for inducing the OCCT. The method was evaluated using data sets from three different domains. The results affirm the effectiveness of the proposed method and show that the OCCT yields better performance in terms of precision and recall (in most cases it is statistically significant) when compared to a C4.5 decision tree-based linkage method.

Credal-C4.5: Decision tree based on imprecise probabilities to classify noisy data

In the area of classification, C4.5 is a known algorithm widely used to design decision trees. In this algorithm, a pruning process is carried out to solve the problem of the over-fitting. A modification of C4.5, called Credal-C4.5, is presented in this paper. This new procedure uses a mathematical theory based on imprecise probabilities, and uncertainty measures. In this way, Credal-C4.5 estimates the probabilities of the features and the class variable by using imprecise probabilities. Besides it uses a new split criterion, called Imprecise Information Gain Ratio, applying uncertainty measures on convex sets of probability distributions (credal sets). In this manner, Credal-C4.5 builds trees for solving classification problems assuming that the training set is not fully reliable. We carried out several experimental studies comparing this new procedure with other ones and we obtain the following principal conclusion: in domains of class noise, Credal-C4.5 obtains smaller trees and better performance than classic C4.5.

An automated approach to quadrilateral mesh generation with complex geometric feature constraints

An automated approach to quadrilateral mesh generation with complex internal geometric feature constraints is presented in this paper. It can deal with all kinds of feature constraints such as internal holes, constraint lines, constraint points, density lines and density points, etc., and satisfy special requirements of mesh generation for numerical analysis. The quadrilateral mesh is generated based on the looping algorithm. As the core of the algorithm, the new splitting criteria are put forward to improve the quality and efficiency of mesh generation. The method of dealing with feature constraints is proposed by considering constraint lines and points, density lines and density points as internal holes with zero area. The method for generating boundary elements is also introduced to improve the element quality around the boundary and feature constraints. For the situation in which feature constraints subdivide the domain into sub-domains, an automatic determination method of sub-domain boundaries is presented. An improved looping algorithm is presented for 3D surface meshing with feature constraints. The determination of proper splitting plane and the handling of feature constraints are put forward. The program of quadrilateral mesh generation has been developed based on the method presented in this paper and successfully applied to mesh generation in several engineering fields.

Ensembles of $({\alpha})$-Trees for Imbalanced Classification Problems

This paper introduces two kinds of decision tree ensembles for imbalanced classification problems, extensively utilizing properties of α-divergence. First, a novel splitting criterion based on α-divergence is shown to generalize several well-known splitting criteria such as those used in C4.5 and CART. When the α-divergence splitting criterion is applied to imbalanced data, one can obtain decision trees that tend to be less correlated (α-diversification) by varying the value of α. This increased diversity in an ensemble of such trees improves AUROC values across a range of minority class priors. The second ensemble uses the same alpha trees as base classifiers, but uses a lift-aware stopping criterion during tree growth. The resultant ensemble produces a set of interpretable rules that provide higher lift values for a given coverage, a property that is much desirable in applications such as direct marketing. Experimental results across many class-imbalanced data sets, including BRFSS, and MIMIC data sets from the medical community and several sets from UCI and KEEL are provided to highlight the effectiveness of the proposed ensembles over a wide range of data distributions and of class imbalance.

Rural parents' perceived stigma of seeking mental health services for their children: development and evaluation of a new instrument.

The purpose of our research was to examine the validity of score interpretations of an instrument developed to measure parents' perceptions of stigma about seeking mental health services for their children. The validity of the score interpretations of the instrument was tested in 2 studies. Study 1 employed confirmatory factor analysis (CFA), using a split half approach, and construct and criterion validity on data from the entire sample of parents in rural Appalachia whose children were experiencing psychosocial concerns (N = 347), while Study 2 employed CFA, construct and criterion validity, and predictive validity of the scores on data from a general sample of parents in rural Appalachia (N = 184). Results of exploratory and confirmatory factor analyses revealed support for a 2-factor model of parents' perceived stigma, which represented both self and public forms of stigma associated with seeking mental health services for their children, and correlated with existing measures of stigma and other psychosocial variables. Further, the new self and public stigma scale significantly predicted parents' willingness to seek services for children.

Straight Line Segments Extraction and EKF-SLAM in Indoor Environment

Thispaper presents a method of simultaneous localization and mapping (SLAM) in indoor environment using extended Kalman filter (EKF) with the straight line segments as the adopted geometrical feature. By using conventional two dimensional laser range finder (LRF) as the main sensor, robot finds a number of points scanned from the surrounding environment.Split-and-Merge is one of the mostpopular algorithms to extract straight line segments from these raw scanned points. However, during the splitting procedure, especially when long range LRF is employed, the widely adoptedmethod named iterative end point fitting (IEPF) has difficulty to segment the points' cluster into collinear subsets correctly and sufficiently due to its constant splitting criterion.To solve this problem,we introduce modified iterative end point fitting (MIEPF) which calculates the splitting criterion for each cluster individually.Simulation and experimental results show the effectiveness of the proposed algorithm. 

Several splitting criteria for vector bundles and reflexive sheaves

Several splitting criteria for vector bundles and reflexive sheaves

A new multi-section based technique for constrained optimization problems with interval-valued objective function

In this paper, an efficient optimization technique is proposed for constrained optimization problems with interval valued objective function. At first, the significance of interval-valued objective function and the meaning of the interval-valued solution of the proposed problem have been explained with graphical interpretation. Generally, this type of problems has infinitely many compromise solutions. The aim of this approach is to obtain one of such solutions with higher accuracy and lower computational cost. The proposed technique is mainly based on the splitting criterion of the accepted/prescribed search region, calculation of the interval inclusion functions and the selection of subregion depending on the modified interval order relations in the context of the decision makers’ point of view. Novel interval oriented constraint satisfaction rules are used for non-interval equality and inequality constraints. Clearly, the proposed technique is nothing but an imitation of well known interval analysis based branch and bound (B &B) optimization approach. The modified multi-section division criterion with some new interval oriented constraint satisfaction rules for non-interval-valued equality and inequality constraints and some novel interval order relations in the context of the decision makers’ point of view have been applied to increase the efficiency of the proposed algorithm. Finally, the technique is applied for solving some test problems and the results are compared with the same obtained from the existing methods.

Study on Splitting Criterion Based on Improvement <i>ucosφ</i>

In this paper, according to the splitting criterion ucosφ, the effectiveness and boundedness of the improvement method is studied. As the oscillation center may be an isolated dot or exists in a section of the splitting, this paper has proposed the thought of false oscillation center and introduced the relative degree coefficient of oscillation k which reflects the degree of effect on the tie lines. The tie lines being cut off should be sorted based on the value of k. If the system recovers stable, the tie lines remained should not be cut off. It is vital to the control of the splitting and reducing the losses of accidents. In the end, the simulation is given in the IEEE 10 Generator 39 Bus System which proves the correctness and effectiveness of this method.

Coronary Heart Event Analysis with Association Rule Mining

Coronary heart disease (CHD) is one of the major causes of disability in adults as well as one of the main causes of death in the developed countries. Although significant progress has been made in the diagnosis and treatment of CHD, further investigation is still needed. The objective of this study was to develop the assessment of heart event-risk factors targeting in the reduction of CHD events using Association Rule Mining. The risk factors investigated were: 1) before the event: a) non modifiable—age, sex, and family history for premature CHD, b) modifiable—smoking before the event, history of hypertension, and history of diabetes; and 2) after the event: modifiable—smoking after the event, systolic blood pressure, diastolic blood pressure, total cholesterol, high density lipoprotein, low-density lipoprotein, triglycerides, and glucose. The events investigated were: myocardial infarction (MI), percutaneous coronary intervention (PCI), and coronary artery bypass graft surgery (CABG).Data-mining analysis was carried out using the Association Rule Mining for the afore mentioned three events using five different splitting criteria for larger datasets. The most important risk factors, as extracted from the classification rules analysis were: 1) for MI, age, smoking, and history of hypertension; 2) for PCI, family history, history of hypertension, and history of diabetes; and 3) for CABG, age, history of hypertension, and smoking. It is anticipated that data mining could help in the identification of high and low risk subgroups of subjects, a decisive factor for the selection of therapy, i.e., medical or surgical.

Random forests on Hadoop for genome-wide association studies of multivariate neuroimaging phenotypes

MotivationMultivariate quantitative traits arise naturally in recent neuroimaging genetics studies, in which both structural and functional variability of the human brain is measured non-invasively through techniques such as magnetic resonance imaging (MRI). There is growing interest in detecting genetic variants associated with such multivariate traits, especially in genome-wide studies. Random forests (RFs) classifiers, which are ensembles of decision trees, are amongst the best performing machine learning algorithms and have been successfully employed for the prioritisation of genetic variants in case-control studies. RFs can also be applied to produce gene rankings in association studies with multivariate quantitative traits, and to estimate genetic similarities measures that are predictive of the trait. However, in studies involving hundreds of thousands of SNPs and high-dimensional traits, a very large ensemble of trees must be inferred from the data in order to obtain reliable rankings, which makes the application of these algorithms computationally prohibitive.ResultsWe have developed a parallel version of the RF algorithm for regression and genetic similarity learning tasks in large-scale population genetic association studies involving multivariate traits, called PaRFR (Parallel Random Forest Regression). Our implementation takes advantage of the MapReduce programming model and is deployed on Hadoop, an open-source software framework that supports data-intensive distributed applications. Notable speed-ups are obtained by introducing a distance-based criterion for node splitting in the tree estimation process. PaRFR has been applied to a genome-wide association study on Alzheimer's disease (AD) in which the quantitative trait consists of a high-dimensional neuroimaging phenotype describing longitudinal changes in the human brain structure. PaRFR provides a ranking of SNPs associated to this trait, and produces pair-wise measures of genetic proximity that can be directly compared to pair-wise measures of phenotypic proximity. Several known AD-related variants have been identified, including APOE4 and TOMM40. We also present experimental evidence supporting the hypothesis of a linear relationship between the number of top-ranked mutated states, or frequent mutation patterns, and an indicator of disease severity.AvailabilityThe Java codes are freely available at http://www2.imperial.ac.uk/~gmontana.

Splitting mixed groups of torsion-free finite rank II

First we introduce the concept of QD-hulls in arbitrary abelian groups. Then we use the concept to give a new characterization of purifiable torsion-free finite rank subgroups of an arbitrary abelian group. Finally we use it to formulate a splitting criterion for mixed groups of torsion-free finite rank.

Tensor products of division algebras and fields

This paper began as an investigation of the question of whether D1⊗FD2 is a domain where the Di are division algebras and F is an algebraically closed field contained in their centers. We present an example where the answer is “no”, and also study the Picard group and Brauer group properties of F1⊗FF2 where the Fi are fields. Finally, as part of our example, we have results about division algebras and Brauer groups over curves. Specifically, we give a splitting criterion for certain Brauer group elements on the product of two curves over F.

Multi-Relational Algebra and its Application to Unrealized Datasets used in C4. 5

Relational databases are based on the theory of relational algebra because all the operations of RDBMS draw their functioning from the operations in relational algebra. The operations of relational algebra are defined on the sets, however, In general, the datamining algorithms requires databases which adopts the multiset philosophy to give better and more accurate results. Unrealized datasets ensures confidentiality of the actual datasets in the datamining process.C4.5 is a classic algorithm which works on mixed real world datasets. This paper proposes the application of Relational algebra for multisets to find the split criterion to be used in classification by the C4.5 algorithm. The results are shown by making the changes in original C4.5 algorithm in the weka tool setting.

Difficult intubation assessment using statistical factor analysis decision tree

Correct and rapid tracheal intubation is an essential anesthesia task for surgical operations. Intubation highly depends on the subjective judgment and experience of the anesthetist. This paper proposes a statistical factor analysis approach to model the preferences of expert anesthetists to enable more accurate pre-operation judgments in cases of difficult intubation. Factor analysis combined with the mutual information between factors is used to generate a robust decision tree (DT) using Bartlett’s node splitting criterion for better decision-making. A tablet computer application is also developed to assist judgment. Several experiments were performed to investigate judgment accuracy and learning effects. Our proposed approach outperformed both a well-known C5.0 DT and an expert opinion derived DT. Encouraging results concerning robustness and efficiency were observed for our approach.

A cohomological splitting criterion for locally free sheaves on arithmetically Cohen–Macaulay surfaces

AbstractIn this paper we obtain a cohomological splitting criterion for locally free sheaves on arithmetically Cohen–Macaulay surfaces with cyclic Picard group, which is similar to Horrocks' splitting criterion for locally free sheaves on projective spaces. We also recover a duality property which identifies a general K3 surface with a certain moduli space of stable sheaves on it, and obtain examples of stable, arithmetically Cohen–Macaulay, locally free sheaves of rank two on general surfaces of degree at least five in ${\mathbb P}^3$.

Self‐Biasing Photoelectrochemical Cell for Spontaneous Overall Water Splitting under Visible‐Light Illumination

A self-biasing photoelectrochemical (PEC) cell that could work for spontaneous overall water splitting in a neutral solution was established based on the mismatched Fermi levels between the photoelectrodes. A Pt-catalyst-decorated crystalline silicon photovoltaic cell (Pt/PVC) was prepared and employed as an effective photocathode. This was coupled with a poly(ethylene glycol)-directed WO3/W photoanode prepared by a hydrothermal process. Both of the photoelectrodes showed a response to visible light. The WO3/W photoanode had a positively located valence band edge, the energy level of which was enough for water oxidation, and the Pt/PVC photocathode possessed a negatively located conduction band edge, which was capable of water reduction. More importantly, the Fermi level of the WO3/W photoanode was more positive than that of the Pt/PVC photocathode because of the p-n junction of the PVC that decoupled the band bending and enlarged the photovoltage. Under visible-light irradiation, the WO3/W photoanode provided a negative bias for the Pt/PVC photocathode, and the Pt/PVC photocathode provided a positive bias for the WO3/W photoanode. An interior bias was generated that could relax the strict criteria of overall water splitting by cooperatively separating the hole-electron pairs at both photoelectrodes. In this system, the short-circuit current and the open-circuit voltage increased with increasing light intensity (AM 1.5 illumination) to reach 121 μA cm(-2) and 0.541 V, respectively, at a light intensity of 100 mW cm(-2). Such a combination provides a promising method for the fabrication of self-driven devices for solar-energy storage.

Payment by episodes of care and comorbidity in Austria: a cross-sectional analysis

BackgroundAustria is currently undergoing a health-care reform, and the issue of integration of care is high on the political agenda. Therefore, payment by episodes of care is a topic of particular interest. This project concentrates on the typical ambulatory services and costs associated with the most important chronic conditions. Since chronic diseases are often combined with other diseases, we specifically try to identify the effect of comorbidity on services and costs. MethodsWe used a full dataset of the Austrian insured population for 2006–07. Robust linear regression was applied to identify the service bundles that are significantly associated with a particular disease. Using these service bundles we calculated disease-specific average ambulatory costs for one episode of care. We accounted for the heterogeneity of needs and costs within the diagnostic groups by identifying split criteria (age and sex). The same methods were applied to assess the effects of comorbidity. FindingsWe obtained typical service bundles for a number of common chronic diseases, to give the average costs for an episode of care. As expected, costs were higher for all comorbidities under research than for their single indications. The cost increase was not trivial. First, average costs of comorbidities were in general not the sum of the costs of their individual diseases, and the cost increase varied greatly by the type of comorbidity. Second, the interquartile range of costs increased greatly for comorbid patients, which means that average costs should be interpreted carefully. InterpretationWe have shown that our data and our methods lead to plausible results for a number of chronic conditions. A special focus should be put on people with comorbidities when developing a payment scheme based on episodes of care. FundingSocial Health Insurance of Austria.