Partitioning Criteria Research Articles

Aggregation of comparisons data and reversal phenomena of metrological interest

Aggregation of comparisons data to rank experimental results and take decisions is being more and more practiced in diverse areas, spanning over a variety of disciplines including, e.g., quality function deployment in industrial engineering, scientometrics, and recovery rate testing of new medications. Problems in decision making may be accrued from the presence of hidden confounding interactions, spurious relationships, lurking variables at work. An analysis of partitioned datasets is carried-on using contingency tables and conditional probabilities. The focus is on intermediate interpretation of evidence to avoid paradoxical reversal of statistical inference when passing from sub-level data to the global level: to this aim, care in partitioning criteria is needed to balance distribution of partitioned data over successive levels, not to incur statistical dependence. An example of counter-intuitive amalgamation effects – also known as Yule-Simpson’s “paradox” – is presented and discussed, showing how to prevent such effects by proper design of experiments.

Context-Aware Hypergraph Construction for Robust Spectral Clustering

Spectral clustering is a powerful tool for unsupervised data analysis. In this paper, we propose a context-aware hypergraph similarity measure (CAHSM), which leads to robust spectral clustering in the case of noisy data. We construct three types of hypergraph---the pairwise hypergraph, the k-nearest-neighbor (kNN) hypergraph, and the high-order over-clustering hypergraph. The pairwise hypergraph captures the pairwise similarity of data points; the kNN hypergraph captures the neighborhood of each point; and the clustering hypergraph encodes high-order contexts within the dataset. By combining the affinity information from these three hypergraphs, the CAHSM algorithm is able to explore the intrinsic topological information of the dataset. Therefore, data clustering using CAHSM tends to be more robust. Considering the intra-cluster compactness and the inter-cluster separability of vertices, we further design a discriminative hypergraph partitioning criterion (DHPC). Using both CAHSM and DHPC, a robust spectral clustering algorithm is developed. Theoretical analysis and experimental evaluation demonstrate the effectiveness and robustness of the proposed algorithm.

浅析P/T界线层中牙形刺Hindeodus parvus的标准性

本文讨论了中国(江西修水东岭、四川广元上寺和浙江长兴煤山)、印度(克什米尔Guryul Ravine)、前苏联(阿塞拜疆Achura)等地的海相P/T界线层中牙形刺Hindeodus parvus，及其首次出现的点位，阐述了其标准性对海相P/T界线划分的重要意义。建议以产于单相且沉积连续剖面，主齿和细齿均保存完好的Hindeodus parvus的首次出现的点位，作为P/T生物地层界线划分的标准。The article discusses the Conodont Hindeodus parvus in marine P/T boundary layer in China (Dongling Village of Xiushui County, Jiangxi Province; Shangsi Village of Guangyuan City, Sichuan Province; Meishan Town of Changxing County, Zhejiang Province), India (Kashmir Guryul Ravine), and Union of Soviet Socialist Republics (Azerbaijan Achura), and the point location where they appear for the first time. It also explains the important significance of setting criterion for demar-cating marine P/T boundary layer, suggesting that the partition criterion for P/T biostratigraphic boundary would be set on the point locations where they firstly appear in single-phase sediment- successive section, with well-preserved cardinal tooth and serration of Hindeodus parvus.

Objective identification study of midsummer rainy season in the northeast China

Owing to the fact that the objective identification methods of rainy season are not totally identical in the three provinces of Northeast China, no unified monitoring criterion of rainy season is available in this area. In order to implement the regional key climate monitoring and to meet the needs of the national weather service, from the angle of the whole region, based on the daily precipitation data of 102 observational stations, by comprehensively analyzing the daily precipitation series disposed by 5-point moving average, the objective identification method of midsummer rainy season in the northeast China is presented in this paper. Then based on the daily reanalysis data of wind field and geopotential height field provided by NCEP/NCAR, the atmospheric circulation fields of different levels before, during and after the midsummer rainy season are contrastively analyzed, and the rationality of the objective identification method is validated. Finally, the partition criterion of the beginning and ending dates are determined. Results show that: the average beginning and ending dates of midsummer rainy season in the northeast China respectively are June 26th and August 30th. The partition criterion of midsummer rainy season in the northeast China by using the objective identification method is reasonable.

Generalization of the MAFRAM Methodology for Semi-Volatile Organic Agro-Chemicals

A wide variety of semi-volatile organic chemicals (SVOCs) are still in use in agricultural practices. A proper understanding of the environmental fate and ecotoxicological risk associated with these compounds can aid decision making, particularly regarding product registration and licensing. The aim of this paper is to expand the use of a previously developed Multimedia Agricultural Fate and Risk Assessment Model (MAFRAM) to SVOCs by adopting the fugacity concept as a second criterion to the existing MAFRAM partitioning criterion (i.e., aquivalence). Volatilization processes from surface compartments into the atmosphere were also included. For example, the application of the generalized model was illustrated using an average annual application rate of 4.48 kg/ha of chlorpyrifos over a typical homogeneous region. Chlorpyrifos emissions were assumed to take place in three environmental compartments (i.e., soil, air, and aboveground plants) with fractions of 0.1, 0.3, and 0.6, respectively. The trends seen in the modeling results were in good agreement with the existing experimental data. Validation issues in MAFRAM were also discussed. Comprehensive experimental validation is unattainable because of the large scale of the areas covered, the lack of boundaries for the system considered, and the uncertainty in the input parameters.

Sparse frontal face image synthesis from an arbitrary profile image

Frontal face image synthesis from an arbitrary profile image plays an important role in automatic video surveillance systems, and remains a challenge in computer vision. The strategies of partition are popular and promising for synthesizing frontal face images. However, conventional rectangular partition criterions fail to align corresponding patches in profile images and frontal face images. Given an arbitrary profile image, to synthesize a corresponding frontal face image which is smooth in texture and similar in appearance, we introduce a triangulation-based partition criterion and do synthesis based on sparse representation. The triangulation-based partition ensures the corresponding triangular patches are strictly aligned. And sparse representation adaptively finds the most similar patches for synthesis while abandons unlike patches. Furthermore, a confederate learning strategy is proposed to reduce the blocking artifacts caused by triangulation-based partition. Experimental results conducted on the Oriental Face database demonstrate the effectiveness of the proposed frontal face image synthesis method and advantages over previous works.

Physical Layer Network Coding for Two-Way Relaying with QAM

The design of modulation schemes for the physical layer network-coded two-way relay network has been studied in literature and it is known that every network coding map that satisfies the exclusive law is representable by a Latin Square. This relationship has been used to get network coding maps satisfying the exclusive law. But, only the scenario in which the end nodes use M-PSK signal sets has been addressed previously. In this paper, we address the case in which the end nodes use M-QAM signal sets. In a fading scenario, for certain channel conditions γe <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">jθ</sup> , termed singular fade states, the end-to-end performance is greatly reduced. By formulating a procedure for finding the exact number of singular fade states for QAM, we show that square QAM signal sets give lesser number of singular fade states compared to PSK signal sets. This reduces the computational complexity at the relay node. It is shown that the criterion for partitioning the complex plane, for the purpose of using a particular network code for a particular fade state, is different from that used for M-PSK. Using a modified criterion, we describe a procedure to analytically partition the complex plane representing the channel condition. We show that when M-QAM (M >4) signal set is used, the conventional XOR network mapping fails to remove the ill effects of γe <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">jθ</sup> =1, which is a singular fade state for all signal sets of arbitrary size. We show that a doubly block circulant Latin Square removes this singular fade state for M-QAM. Finally it is shown that M-QAM gives superior performance over M-PSK.

An Optimization Designing Method of Fractionated Spacecraft

This article systematically illustrates the concept of the fractionated spacecraft and analyzes the project evaluation and the optimized designs of fractionated spacecraft system from domestic and foreign scholars, then based on those, they presented a new index that random task cycle costs should be regarded as a comprehensive assessment criterion. Next, through modeling the uncertain cost of module development, launch/deployment and full task cycle of Monte Carlo simulation, it provides random task cycle cost of every configuration scheme. Lastly, it establishes the partitioning criteria and partitioning method system for fractionated spacecraft, with further modules' partitioning road map based on pedigree clustering method. The optimized designs are determined according to classification level combined with the actual situations.

Reference interval for lipid profile in North Indian population from Rajasthan according to various partitioning criteria

ObjectiveLipid profile parameters are influenced by various factors like age, ethnicity, diet, genetic and gender differences hence it is essential to establish reference range of the values of serum lipids for a given population in India. We have planned this study to evaluate the reference values of lipid profile of a North Indian population according to the guidelines of the National Cholesterol Education Program (NCEP) of the USA. Design and methodsThe present study was conducted on 2021 apparently healthy individuals of North Indian origin ranging in age from 15 to 60years, who were selected randomly using defined criteria. Fasting samples were analyzed for total cholesterol, triglyceride, HDL-C and LDL-C. Data were analyzed for middle 95th percentile (2.5th–97.5th percentile), median and 95% confidence interval using SPSS software package version 10.0. ResultsNo substantial difference could be observed between male and female and vegetarian and non-vegetarian, in cholesterol, triglyceride and LDL-C levels. However HDL-C reported higher limit in female as compared to male (33–64 vs 32–58mg/dl). Similarly upper limit of HDL-C in vegetarians were higher than non-vegetarian (value 32.8–64.92 vs 30.72–58.10mg/dl). Median value for cholesterol, triglyceride, LDL-C progressively increased in different age groups (<20, 20–40 and 41–60years). No marked difference was observed in reference interval of these parameters in rural and urban populations. ConclusionIt can be suggested that lipid values obtained in this study can be used as the reference value, based on which clinical correlation can be made.

A distributed implementation of multi-area power system state estimation on a cluster of computers

This paper presents an efficient weighted least squares (WLS) distributed algorithm for multi-area power system state estimation including measurements provided by the supervisory control and data acquisition (SCADA) system and the synchronized phasor measurement units (PMUs). Each area performs independently its own state estimation using local measurements and exchanges border information at a coordination level that computes the system-wide state. The basic criterion of partitioning a power system into several control areas is to have areas as equal in size as possible, so that the workload on each area processor is as balanced as possible, and interconnections between distinct areas be limited, as much as possible, to reduce the amount of inter-process communication necessary. A true distributed implementation on a cluster of computers running Linux is developed. The code is written in C and sparse linear algebra routines from the Portable, Extensible Toolkit for Scientific computation (PETSc) library are used. Communication among area processors and the coordinator processor is implemented by using the MPICH2 implementation of the message-passing interface (MPI) protocol. By using extensions MPICH-G2, ES-MPICH2, MVAPICH2 of MPICH2 protocol, a portable Web-based distributed multi-area state estimator can be developed based on geographically distributed and heterogeneous cluster of computers. The proposed distributed approach has the same accuracy and redundancy level as the centralized estimator and provides the same solution only from the strictly mathematical point-of-view, by assuming that the measurement arrival delays and skew errors are identical to those of the centralized state estimator. Numerical simulations with a 1180 bus system, partitioned into equally or unequally sized area networks, demonstrate the efficacy of the proposed implementation in comparison with a centralized WLS state estimation.

Multiclass classifier based on boundary complexity

This paper presents several criteria for partition of classes for the support vector machine based hierarchical classification. Our clustering algorithm combines support vector machine and binary tree, it is a divisive (top-down) approach in which a set of classes is automatically separated into two smaller groups at each node of the hierarchy, it splits the classes based on the normalized cuts clustering algorithm. Our clustering algorithm considers the involved classes rather than the individual data samples. In the new proposed measures, similarity between classes is determined based on boundary complexity. In these measures, concepts such as the upper bound of error and Kolmogorov complexity are used. We reported results on several data sets and five distance/similarity measures. Experimental results demonstrate the superiority of the proposed measures compared to other measures; even when applied to nonlinearly separable data, the new criteria perform well.

Black-Start Subsystem Partitioning Based on Nodes Voltage

Divide the network into several subsystems then restore the subsystems separately can accelerate the restoration procedure after a main blackout occurs. In this paper, the strategy of black-start system partitioning based on nodes voltage is proposed. The strategy considers the node voltage as partition criterion. By power flow calculating gets nodes voltage and sorts them and then select voltage threshold to partitioning the groups of the nodes according to nodes voltage and the number of black-start units. The algorithm has a rigorous theoretical basis for it based on communities finding theory of complex network. The effectiveness of the proposed method is validated by studying the IEEE 118-bus case.

A Model for Pairs of Beatty Sequences

Beatty sequences ⌊nα1 + ϐ1※m∈ℤ and ⌊mα2 + ϐ2※m∈ℤ are described by two athletes running in opposite directions on a round track. This approach suggests an interesting interpretation for well-known partitioning criteria; such sequences (eventually) partition the integers essentially when the athletes have the same starting point.

Partition dataset according to amino acid type improves the prediction of deleterious non-synonymous SNPs

Many non-synonymous SNPs (nsSNPs) are associated with diseases, and numerous machine learning methods have been applied to train classifiers for sorting disease-associated nsSNPs from neutral ones. The continuously accumulated nsSNP data allows us to further explore better prediction approaches. In this work, we partitioned the training data into 20 subsets according to either original or substituted amino acid type at the nsSNP site. Using support vector machine (SVM), training classification models on each subset resulted in an overall accuracy of 76.3% or 74.9% depending on the two different partition criteria, while training on the whole dataset obtained an accuracy of only 72.6%. Moreover, the dataset was also randomly divided into 20 subsets, but the corresponding accuracy was only 73.2%. Our results demonstrated that partitioning the whole training dataset into subsets properly, i.e., according to the residue type at the nsSNP site, will improve the performance of the trained classifiers significantly, which should be valuable in developing better tools for predicting the disease-association of nsSNPs.

Resource Allocation with Partitioning Criterion for Macro-Femto Overlay Cellular Networks with Fractional Frequency Reuse

The interference mitigation technique based on fractional frequency reuse (FFR) provides improved cell-edge performance with similar overall cell capacity as that of systems with the frequency reuse factor of one. Furthermore, frequency sub-band allocation by FFR has the benefit of allowing flexibility for the deployment of femto-cells through frequency partitioning. Determination of a proper frequency partitioning criterion between the cell-center and the cell-edge, and between the cells with femto-cells is an important issue. In addition, time resource partitioning introduces another degree of freedom to the design of time-frequency resource allocation. In this paper, we propose a novel time-frequency resource allocation mechanism using FFR for a macro-femto overlay cellular network. Feasible frequency sub-band and time resource is allocated to the cell-center and the cell-edge region in a cell by the proposed partitioning criterion and the time partitioning ratio. We provide a guideline for how to determine the partitioning criterion for the regions and how to design the amount of time resource. We derive the average capacity of macro-cells and femto-cells, and introduce a new harmonic mean metric to maximize the average capacity of the regions while achieving the fairness among users in a cell.

Reversible Image Data Hiding Using Quad-tree Segmentation and Histogram Shifting

This paper is working towards a high capacity histogram-based reversible data hiding algorithm with a relatively lower distortion introduced after embedding the secret message. The proposed algorithm can be thought of as an improved version of the traditional histogram shifting method. To take advantage of simple local distribution of pixel intensities, the idea behind the proposed algorithm is to consider the multiple local histograms, each of which is derived from pixels in a locally contiguous sub-image area, instead of the single histogram for the whole image area. We develop a hierarchical segmentation scheme that is able to hierarchically partition the input host image into several variable-sized blocks of pixels under a maximal capacity criterion for each block partition. These partitioned blocks are organized as a tree structure for the ease of representation. The secret message and the partition tree information are then embedded in these image blocks. With the proposed segmentation scheme, the algorithm can easily find a suitable non-overlapped partition of the image to significantly increase embedding capacity. Experimental results reveal that the proposed method can indeed provide higher embedding capacity than the traditional histogram shifting method while maintaining a high visual quality that is comparable to the traditional one without spatial segmentation scheme.

An Improved Segmentation Algorithm for Infrared Imaging Object

In this paper, an improved threshold segmentation algorithm based on isoperimetric ratio for infrared imaging object is proposed. The segmentation weight matrix is constructed by computing the similarity among the pixies with 4 adjoining points and stored in a sparse matrix. The isoperimetric ratio is obtained after the indicator vectors are formed with 255 gray levels. The proposed algorithm selects the minimum isoperimetric ratio confined in conditions as the best partition criterion instead of the traditional minimum isoperimetric ratio. By analyzing the variation of isoperimetric ratio with the gray levels, the proposed method can find the optimum threshold to segment infrared imaging object. Experimental results show that compared with the traditional methods, the proposed algorithm can reach a higher segmentation rate and is more robust in different kinds of infrared images.

Vehicle Routing Problem with Time Windows Based on Spatiotemporal Distance

Vehicle Routing Problem with Time Windows (VRPTW) is a well known NP-hard problem. A typical way to solve this problem is to partition the customers into groups first and then construct optimal routes for each group or groups. Conventionally, customers are partitioned according to the spatial distance only, while ignoring their time window requirements. In this paper, we consider jointly spatial and temporal characteristics of customers and propose a spatiotemporal distance based criteria for customers partitioning. To improve the initial solution, a tabu search algorithm is developed. To improve the convergence efficiency, this algorithm limits the objective value ranges that around the recently visited solutions, Instead of limiting recently visited solutions, the performance of the proposed algorithm is examined with 56 Solomon benchmark problems. The results indicate that spatiotemporal distance is more effective than spatial distance when solving vehicle routing problems with narrow time windows. Some results are even comparable to the best results obtained so far.

Main partitioning criteria for the characterization of the health status in the freshwater mussel Anodonta cygnea from spontaneously polluted area in western ukraine

The aim of this study was to appreciate the consequences of spontaneous human activity for freshwater mollusks in the generally ecologically sustainable area in Western Ukraine. For this, bivalve mollusk, Anodonta cygnea, at three sites, with mixed agricultural and municipal activities (A), close to a municipal water inlet (F) and the cooling pond of a nuclear power plant (N), were studied in spring, summer, and autumn. The set of parameters included the characteristics of oxidative stress (activity of catalase (CAT), levels of protein carbonyls (PC)), levels of reduced and oxidized glutathione (GSH, GSSG, respectively), activities of lactate dehydrogenase (LD), cholinesterase (ChE), ethoxyresorufin-O-deethylase (EROD), glutathione S-transferase (GST) in the digestive gland, and concentrations of vitellogenin-like proteins (Vtg-LP) in gonads and also morphological indices. Although the discriminant functional analysis confirmed the general seasonal regularities for studied groups, it allowed to discriminate between sites (P < 0.05). At site A, oxidative stress; high levels of LD, EROD, and GST; and low levels of ChE and condition factor were reflected. This demonstrated the sensitivity of mussels to constant effect of mixed pollution. At site N, oxidative injury was shown that might be explained by the constantly high temperature. At site F, abrupt elevations of Vtg-LP and EROD levels in autumn were probably related to an emergency situation on the nearby dump. So, both chronic and temporal environmental effects were reflected by a set of markers in mollusk. The classification and regression tree (CART) algorithm selected GSH and PC in the digestive gland and Vtg-LP as partitioning criteria for the characterization of mussel health status. © 2010 Wiley Periodicals, Inc. Environ Toxicol, 2012.

Developing seed zones and transfer guidelines with multivariate regression trees

Managing seed movement is an important component of forest resource management to minimize maladaptation of planting stock in forest plantations. Here, we describe a new approach to analyze geographic patterns of adaptive and neutral genetic variation in forest trees and to link this genetic information to geographic variables for the delineation of seed zones and the development of seed transfer guidelines. We apply multivariate regression trees to partition genetic variation, using a set of environmental or geographic predictor variables as partitioning criteria in a series of dichotomous splits of the genetic dataset. The method can be applied to any type of genetic data (growth, adaptive, or marker traits) and can simultaneously evaluate multiple traits observed over several environments. The predictor variables can be categorical (e.g., ecosystem of seed source), continuous (e.g., geographic or climate variables), or a combination of both. Different sets of predictor variables can be used for different purposes: In two case studies for aspen and red alder, we show (1) how latitude, longitude, and elevation of seed sources in a provenance trial can be used to develop simple seed transfer guidelines; (2) how ecosystem classes and elevation as predictor variables can be used to delineate seed zones and breeding regions; and (3) how climate variables as predictors can reveal adaptation of genotypes to the environments in which they occur. Partitioning of genetic variation appears very robust regarding the choice of predictor variables, and we find that the method is a powerful aid for interpreting complex genetic datasets.