Computational Science – ICCS 2008

This short paper is intended to give our readers a brief insight about the Seventh International Workshop on Computer Graphics and Geometric Modeling-CGGM’2008, held in Krakow (Poland), June 23-25 2008 as a part of the ICCS’2008 general conference.

Virtual environments for interactive applications demand highly realistic scenarios, which tend to be large and densely populated with very detailed meshes. Despite the outstanding evolution of graphics hardware, current GPUs are still not capable of managing these vast amounts of geometry. A solution to overcome this problem is the use of level-of-detail techniques, which recently have been oriented towards the exploitation of GPUs. Nevertheless, although some solutions present very good results, they are usually based on complex data structures and algorithms. We thus propose a new multiresolution model based on triangles which is simple and efficient. The main idea is to modify the list of vertices when changing to a new level of detail, in contrast to previous models which modify the index list, which simplifies the extraction process. This feature also provides a perfect framework for adapting the algorithm to work completely on the GPU.

We propose a method for efficient calculation of proximity queries for a moving object. The proposed method performs continuous collision detection between two given configurations according to the exact collision checking (ECC) approach which performs distance calculation between two objects. This method obtains efficient results as it employs the concept of clearance bounds and performs approximate distance calculations with a tight fit of bounding volumes. The high efficiency of the method, when applied to robot path planning, is demonstrated through some experiments.

This paper presents a method for fusing multifocus images into enhanced depth-of-field composite image and creating a 3D model of a photographed scene. A set of images of the same scene is taken from a typical digital camera with macro lenses with different depth-of-field. The method employs convolution and morphological filters to designate sharp regions in this set of images and combine them together into an image where all regions are properly focused. The presented method consists of several phases including: image registration, height map creation, image reconstruction and final 3D scene reconstruction. In result a 3D model of the photographed object is created.

The authors have been developing

KETpic

as a bundle of macro packages for Computer Algebra Systems (CASs) to draw fine

${\rm\kern-.15em T\kern-.1667em\lower.7ex\hbox{E}\kern-.125emX}$

-pictures. Recently we have developed a new method of the surface drawing using KETpic. The equation of envelopes is used to draw ridgelines of surfaces. Also the technique of hidden line elimination is used. By these methods, we can draw 3D-graphics which are so simple that the global (i.e. sketchy) shapes of them are easily understood.

We present an improved reflection model optimized for global illumination. The model produces visually plausible images, is symmetric and has improved energy preserving capabilities compared to previous approaches which satisfies these requirements. Having an efficient sampling routine, the model is ready to use in Monte Carlo rendering. Presented model is phenomenological, i.e. it has intuitive glossiness parameter that affects its appearance. Moreover it can be used as a set of basis functions designed to fit material reflection to measured data.

We shall describe a mesh generation technique on a closed CAD surface composed of a few parametric surfaces. The edge size function is a fundamental entity in order to be able to apply the process of generalized Delaunay triangulation with respect to the first fundamental form. Unfortunately, the edge size function is not known a-priori in general. We describe an approach which invokes the Laplace-Beltrami operator to determine it. We will discuss theoretically the functionality of our methods. Our approach is illustrated by numerical results from the harmonicity of triangulations of some CAD objects. The IGES format is used in order to acquire the initial geometries.

This paper presents a method to create sharp features such as creases or corners on non-regular triangular meshes. To represent sharp features on a triangular spline surface we have studied a method that enables designers to control the sharpness of the feature parametrically. Extended meshes are placed to make parallelograms, and then we have an extended vertex which is used to compute control points for a triangular Bézier patch. This extended vertex expressed with a parameter enables designers to change the shape of the sharp features. The former method we presented deals with regular meshes, however, it can be a strong restriction against the actual variety of meshes. Therefore, we developed a method to express sharp features around an extraordinary vertex. In this paper, we present algorithms to express creases and corners for a triangular mesh including extraordinary vertices.

Art often provides valuable hints for technological innovations especially in the field of Image Processing and Computer Graphics. In this paper we present a novel method to generate an artificial mosaic starting from a raster input image. This approach, based on Gradient Vector Flow computation and some smart heuristics, permit us to follow the most important edges maintaining at the same time high frequency details. Several examples and comparisons with other recent mosaic generation approaches show the effectiveness of our technique.

Applications such as video games or movies often contain deforming meshes. The most-commonly used representation of these types of meshes consists in dense polygonal models. Such a large amount of geometry can be efficiently managed by applying level-of-detail techniques and specific solutions have been developed in this field. However, these solutions do not offer a high performance in real-time applications. We thus introduce a multiresolution scheme for deforming meshes. It enables us to obtain different approximations over all the frames of an animation. Moreover, we provide an efficient connectivity coding by means of triangle strips as well as a flexible framework adapted to the GPU pipeline. Our approach enables real-time performance and, at the same time, provides accurate approximations.

We give a method to generate polynomial approximations to constant mean curvature surfaces with prescribed boundary. We address this problem by finding triangular Bézier extremals of the CMC-functional among all polynomial surfaces with a prescribed boundary. Moreover, we analyze the

$\mathcal{C}^1$

problem, we give a procedure to obtain solutions once the tangent planes for the boundary curves are also given.

Today’s virtual worlds challenge the capacity of human creation. Trying to reproduce natural scenes, with large and complex models, involves reproducing their inherent complexity and detail. Procedural generation helps by allowing artists to create and generalize objects for highly detailed scenes. But existing procedural algorithms can not always be applied to existing applications without major changes. We introduce a new system that helps include procedural generation into existing modeling and rendering applications. Due to its design, extensibility and comprehensive interface, our system can handle user’s objects to create and improve applications with procedural generation of content. We demonstrate this and show how our system can generate both models and behaviours for a typical graphics application.

This work concerns the issue of surface reconstruction, that is, the generation of a surface from a given cloud of data points. Our approach is based on a metaheuristic algorithm, the so-called Particle Swarm Optimization. The paper describes its application to the case of Bézier surface reconstruction, for which the problem of obtaining a suitable parameterization of the data points has to be properly addressed. A simple but illustrative example is used to discuss the performance of the proposed method. An empirical discussion about the choice of the social and cognitive parameters for the PSO algorithm is also given.

In a wide range of industrial applications there appear systems of hard particles of different shapes and sizes, known as “packings”. In this work, the force-biased algorithm, primarily designed to model close packings of equal spheres, is adapted to simulate mixtures of spherocylindrical particles of different radii and aspect ratios. The packing densities of simulated mono and polydisperse systems, are presented as functions of particle elongation and different algorithm parameters. It is shown that spherocylinders can pack more densely than spheres, reaching volume fraction as high as 0.705.

This article presents a new method for foliage illumination which takes into account direct, indirect illumination and self-shadowing. Both indirect illumination and self-shadowing are approximated by means of a novel technique using depth maps. In addition, a new shadow casting algorithm is developed to render shadows produced by the foliage onto regular surfaces which enhances the appeareance of this kind of shadows compared to traditional shadow mapping techniques.

Triangulation techniques along with laser technology have been widely used for 3D scanning, however, 3D scanning of moving object and on-line modeling have received less attention. The current work describes a system developed for on-line CAD model generation using scanned data. The system developed uses structured laser light pattern and a digital CCD camera to generate and capture images from which the range data is extracted. The data is then analyzed to reconstruct the geometric model of the moving object. To exploit the potential of a geometric modeler the model is represented in a commercial CAD software. To reduce errors the system employs on-line calibration, light distribution correction algorithm, camera calibration, and subpixeling techniques. A data reduction scheme has also been incorporated to eliminate redundant data.

In this work an image pre-processing module has been developed to extract quantitative information from plantation images with various degrees of infestation. Four filters comprise this module: the first one acts on smoothness of the image, the second one removes image background enhancing plants leaves, the third filter removes isolated dots not removed by the previous filter, and the fourth one is used to highlight leaves’ edges. At first the filters were tested with MATLAB, for a quick visual feedback of the filters’ behavior. Then the filters were implemented in the C programming language. At last, the module as been coded in VHDL for the implementation on a Stratix II family FPGA. Tests were run and the results are shown in this paper.

Modeling and

Simulation of Multiphysics Multiscale Systems

(SMMS) poses a grand challenge to computational science. To adequately simulate numerous intertwined processes characterized by different spatial and temporal scales spanning many orders of magnitude, sophisticated models and advanced computational techniques are required. The aim of the SMMS workshop is to encourage and review the progress in this multidisciplinary research field. This short paper describes the scope of the workshop and gives pointers to the papers reflecting the latest developments in the field.

We present a computational model of tumor induced sprouting angiogenesis that involves a novel coupling of particle-continuum descriptions. The present 3D model of sprouting angiogenesis accounts for the effect of the extracellular matrix on capillary growth and considers both soluble and matrix-bound growth factors. The results of the simulations emphasize the role of the extracellular matrix and the different VEGF isoforms on branching behavior and the morphology of generated vascular networks.

We discuss a novel metaphor of tumor progression stimulated by the process of angiogenesis. The realistic 3-D dynamics of the entire system consisting of the tumor, tissue cells, blood vessels and blood flow can be reproduced by using interacting particles. The particles mimic the clusters of tumor cells. They interact with their closest neighbors via semi-harmonic forces simulating mechanical resistance of the cell walls and the external pressure. The particle dynamics is governed by both the Newtonian laws of motion and the rules of cell life-cycle. The particles replicate by a simple mechanism of division, similar to that of a single cell reproduction and die due to necrosis or apoptosis. We conclude that this concept can serve as a general framework for designing advanced multi-scale models of tumor dynamics. In respect to spatio-temporal scale, the interactions between particles can define e.g., cluster-to-cluster, cell-to-cell, red blood cells and fluid particles interactions, cytokines motion, etc. Consequently, they influence the macroscopic dynamics of the particle ensembles in various sub-scales ranging from diffusion of cytokines, blood flow up to growth of tumor and vascular network expansion.

We present a first attempt to create an in-silico model of a uterine leiomyoma, a typical exponent of a common benign tumor. We employ a finite element model to investigate the interaction between a chemically driven growth of the pathology and the mechanical response of the surrounding healthy tissue. The model includes neoplastic tissue growth, oxygen and growth factor transport as well as angiogenic sprouting. Neovascularisation is addressed implicitly by modeling proliferation of endothelial cells and their migration up the gradient of the angiogenic growth factor, produced in hypoxic regions of the tumor. The response of the surrounding healthy tissue in our model is that of a viscoelastic material, whereby a stress exerted by expanding neoplasm is slowly dissipated. By incorporating the interplay of four underlying processes we are able to explain experimental findings on the pathology’s phenotype. The model has a potential to become a computer simulation tool to study various growing conditions and treatment strategies and to predict post-treatment conditions of a benign tumor.

Carbon nanotubes (CNTs) grown in a thin film have shown great potential as cathodes for the development several field emission devices. However, in modeling these important devices we face substantial challenges since the CNTs in a thin film undergo complex dynamics during field emission, which includes processes such as (1) evolution, (2) electromechanical interaction, (3) thermoelectric heating and (4) ballistic transport. These processes are coupled, nonlinear, and multiphysics in their nature. Therefore, they must be analyzed accurately from the stability and long-term performance view-point of the device. Fairly detailed physics-based models of CNTs considering some of these aspects have recently been reported by us. In this paper, we extend these models and focus on their computational implementation. All components of models are integrated at the computational level in a systematic manner in order to accurately calculate main characteristics such as the device current, which are particularly important for stable performance of CNT thin film cathodes in x-ray devices for precision biomedical instrumentation. The numerical simulations reported in this paper are able to reproduce several experimentally observed phenomena, which include fluctuating field emission current, deflected CNT tips and the heating process.

We present MUSE, a software framework for tying together existing computational tools for different astrophysical domains into a single multiphysics, multiscale workload. MUSE facilitates the coupling of existing codes written in different languages by providing inter-language tools and by specifying an interface between each module and the framework that represents a balance between generality and computational efficiency. This approach allows scientists to use combinations of codes to solve highly-coupled problems without the need to write new codes for other domains or significantly alter their existing codes. MUSE currently incorporates the domains of stellar dynamics, stellar evolution and stellar hydrodynamics for a generalized stellar systems workload. MUSE has now reached a “Noah’s Ark” milestone, with two available numerical solvers for each domain. MUSE can treat small stellar associations, galaxies and everything in between, including planetary systems, dense stellar clusters and galactic nuclei. Here we demonstrate an examples calculated with MUSE: the merger of two galaxies. In addition we demonstrate the working of MUSE on a distributed computer. The current MUSE code base is publicly available as open source at http://muse.li.

In this paper we present a High Level Architecture (HLA) component model, particularly suitable for distributed multiscale simulations. We also present a preliminary implementation of HLA components and the CompoHLA environment that supports setting up and managing multiscale simulations built in the described model. We propose to integrate solutions from High Level Architecture (such as advanced time and data management) with possibilities given by component technologies (such as reusability and composability) and the Grid (such as joining geographically distributed communities of scientists). This approach will allow users working on multiscale applications to more easily exchange and join the simulations already created. The particular focus of this paper is on the design of a HLA component. We show how to insert simulation logic into a component and make it possible to steer from outside its connections with other components. Its functionality is shown through example of multiscale simulation of dense stellar system.

The ability to couple distinct computational models of science and engineering systems is still a recurring challenge when developing multiphysics applications.

The applied coupling technique is often dictated by various constraints (such as hard- and software requirements for the submodels to be coupled). This may lead to different coupling strategies/implementations in case a submodel has to be replaced in an existing coupled setup.

Additional efforts are required when it comes to multiscale coupling. At least one of the submodels has to be modified to provide a matching interface on a specific spatial and temporal scale.

In the present paper we describe a generic coupling mechanism/framework to reduce these common problems and to facilitate the development of multiscale simulations consisting of a multitude of submodels.

The resulting implementation allows the coupling of legacy as well as dedicated codes with only minor adjustments. As the system is being build upon the JADE library, our platform fully supports computations on distributed heterogeneous hardware.

We discuss the platform’s capabilities by demonstrating the coupling of several cellular-automata kernels to model a coupled transport problem.

We present multiscale simulations of liquid water using a novel control algorithm to couple non-periodic molecular dynamics (MD) and continuum descriptions in the context of a Schwarz alternating method. In the present multiscale approach the non-periodic MD simulations are enhanced by an effective external boundary force that accounts for the virial component of the pressure and eliminates spurious density oscillations close to the boundary. This force is determined by a simple control algorithm that enables coupling of the atomistic description to a coarse grained or a continuum description of liquid water. The proposed computational method is validated in the case of equilibrium and parallel flow.

NASA future exploration missions and space electronic equipment require improvements in solar cell efficiency and radiation hardness. Novel nano-engineered materials and quantum-dot (QD) based photovoltaic devices promise to deliver more efficient, lightweight, radiation hardened solar cells and arrays, which will be of high value for the long term space missions. We describe the multiscale approach to the development of Technology Computer Aided Design (TCAD) simulation software tools for QD-based semiconductor devices, which is based on the drift – diffusion and hydrodynamic models, combined with the quantum-mechanical models for the QD solar cells.

A stationary hot gas jet supersonically expanding into a low pressure environment is studied through multi-scale numerical simulations.

A hybrid continuum-molecular approach is used to model the flow. Due to the low pressure and high thermodynamic gradients, the accuracy of continuum mechanics results are doubtful, while, because of its excessive time expenses, a full molecular method is not feasible. The results of the proposed hybrid continuum-molecular approach have been successfully validated against experimental data.

An important question for the full understanding of the processes governing the flow is addressed: the demonstration of an invasion of the supersonic part of the flow by background particles. Through the tracking of particles and collisions in the supersonic region it could be definitively proven that background particles are present in this region. We present a complete two dimensional picture of how the invading background particles distribute and collide with local particles into the supersonic region.

Multiphysics and multiscale three-phase flow simulation is proposed for model based control. Three-phase flow is considered by means of particles movement in a pipe with two-phase gas and liquid vacuum pumping. The presented model and simulation algorithm were implemented using a software system working in real-time mode. The software system can simulate a part of the pipe net with configured pipe profile, pump station and valve parameters and also inlet mixture composition. In addition, the system includes algorithm for pressure control.

The sound emitted from splash of water droplet colliding with shallow water is simulated by the finite difference lattice Boltzmann method. Two-particle immiscible fluid model is used, and the under water sound is considered by introducing the elasticity for the liquid phase. After the collision, sounds propagating into the gas and liquid phases are successively detected. The directivity of the sound is shown to depend on the depth of the water.

This paper describes the new CFDRC mixed-mode simulator which combines multiscale 3D Technology Computer Aided Design (TCAD) device models (fluid carrier transport and nuclear ion track impact), and advanced compact transistors models. Key features include an interface and 3D adaptive meshing to allow simulations of single event radiation effects with nuclear reactions and secondary particles computed by Vanderbilt’s MRED/Geant4 tools.

Complex Automata (CxA) have been recently proposed as a paradigm for the simulation of multiscale systems. A CxA model is constructed decomposing a multiscale process into single scale sub-models, each simulated using a Cellular Automata algorithm, interacting across the scales via appropriate coupling templates. Focusing on a reaction-diffusion system, we introduce a mathematical framework for CxA modeling. We aim at the identification of error sources in the modeling stages, investigating in particular how the errors depend upon scale separation. Theoretical error estimates will be presented and numerically validated on a simple benchmark, based on a periodic reaction-diffusion problem solved via multistep lattice Boltzmann method.

Multiscale schemes for transferring information from fine to coarse scales are typically based on some sort of averaging. Such schemes smooth the fine scale features of the underlying fields, thus altering the fine scale correlations. As a superior alternative to averaging, a wavelet based scheme for the exchange of information between a reactive and diffusive field in the context of multiscale reaction-diffusion problems is proposed and analyzed. The scheme is shown to be efficient in passing information along scales, from fine to coarse, i.e. up-scaling as well as from coarse to fine, i.e. down-scaling. In addition, it retains fine scale statistics, mainly due to the capability of wavelets to represent fields hierarchically. Critical to the success of the scheme is the identification of dominant scales containing the majority of useful information. The scheme is applied in detail to the analysis of a diffusive system with chemically reacting boundary. Reactions are simulated using kinetic Monte Carlo (KMC) and diffusion is solved by finite differences. Spatial scale differences are present at the interface of the KMC sites and the diffusion grid. The computational efficiency of the scheme is compared to results obtained by local averaging, and to results from a benchmark model. The spatial scaling scheme ties to wavelet based schemes for temporal scaling, presented elsewhere by the authors.

In the case of strongly coupled problems like fluid-structure models in aero-elasticity or aero-thermo-mechanics, a standard solution methodology is based on so called Dirichlet-Neumann iterations. This means that, for instance, the velocity at the interface between the two media is imposed in the fluid, the solution of the fluid problem gives a pressure that is imposed at the boundary of the structure, and then the solution of the problem in the structure gives a new velocity to be imposed to the fluid. This method is not always stable, depending on the relative properties of the media, unless a suitable relaxation parameter is introduced. In order to enforce both velocity and pressure continuity at the interface, the matching conditions can be formulated, like in domain decomposition methods, in a mixed form. This means that the boundary conditions derived in one physical domain from the other one is of Robin type. With Robin boundary condition, an interface stiffness, in the case of velocity-pressure conditions, is introduced. The optimal choice for this stiffness can be proved to be, in the case of linear problems, the so called ”Dirichlet-Neumann” operator of the opposite domain, this means for the discrete equations, the static condensation on the interface of the domain stiffness matrix. Of course, the static condensation cannot be performed in practice, since it is extremely expensive and that the resulting matrix is dense. But it can be approximated in several ways. The underlying general idea behind that methodology is the following: with Robin boundary conditions on the interface, a constitutive law is imposed on the boundary of each media that should optimally exactly represent the interaction with the other media.

We discuss the modeling of transient wave propagation with the boundary element method (BEM) in three dimensions. The special structure of the fundamental solution of the wave equation leads to a close interaction of space and time variables in a so-called retarded time-argument. We give a detailed derivation of the discretization scheme and analyse a new kind of ”geometrical light cone” singularity of the retarded potential function. Moreover, we present numerical experiments that show these singularities.

This paper is devoted to the numerical solution of the instationary Maxwell equations in waveguides with metallic conical inclusions on its internal boundary. These conical protuberances are geometrical singularities that generate in their neighborhood, strong electromagnetic fields. Using some recent theoretical and practical results on curl-free singular fields, we have built a method which allows to compute the instationary electromagnetic field. It is based on a splitting of the spaces of solutions into a regular part and a singular one. The singular part is computed with the help of a multiscale representation, written in the vicinity of the geometrical singularities. As an illustration, numerical results in a rectangular waveguide are shown.

The 3

rd

workshop on computational chemistry and its applications (3

rd

CCA) will be held, as part of ICCS-2008, from 23

rd

to 25

th

June 2008 in Kraków, Poland. This workshop is the third one after being two successful events in Reading, UK and Beijing, China for ICCS-2006 and ICCS-2007 respectively. The main aim of this workshop is to consider only high standards, original work not published, after peer reviewing.

Computational chemistry is one of the fields of computational science and hence this workshop is suitable for the ICCS conference. Computational chemistry is rapidly expanding with the explosive growth of computational power. It is widely used in research and more interestingly in interdisciplinary research involving computational science.

The trans and gauche rotamers of 1,2-diisocyanoethane, 1,2-diisocyanodisilane and isocyano(isocyanomethyl)silane have been studied in the gas phase. A transition state has also been obtained for the interconversion of these rotamers. The methods used are MP2 and DFT/B3LYP and the basis sets used for all atoms are 6-311++G(d,p). The optimised geometries, dipole moments, moment of inertia, energies, energy differences and rotational barriers are reported. Vibrational frequencies of the rotamers are presented with appropriate assignments. The results indicate that the trans rotamer is more stable and the G2MP2 rotational energy differences are 2.97 kJ/mol (1,2-diisocyanoethane), 3.02 kJ/mol (isocyano(isocyanomethyl)silane) and 2.12 kJ/mol (1,2-diisocyanodisilane). The rotational barrier for 1,2-diisocyanoethane is larger than its energy difference but the barrier becomes comparable to the energy difference for isocyano(isocyanomethyl)silane.

The formation of superoxide at Ag(100) and Ag(111) surfaces for cluster and periodic slab models is studied by applying first-principles density functional theory calculations, including ab-initio molecular dynamics. Adsorption energies and structural parameters are discussed in detail. Charge transfer analyses indicate that O2- preferentially forms on clusters, particularly at an Ag(100) surface.

Binding of blockers to the Influenza A ion-channel is studied using automated docking calculations. Our study suggests that studied cage compounds inhibit the M2 ion channel by binding to the His37 residue. The adamantane cage fits into a pocket formed by Trp41 residue, while the hydrogen bond is formed between hydrogen atom of ammonium nitrogen and the nitrogen of histidine residue. This finding is supported by experimental data and should help to obtain better understanding of the inhibition mechanism of the Influenza A M2 ion channel.

This paper describes a computational chemistry solution for the management of large chemical databases of molecules and the performing of isomorphism calculation for the analysis of database similarity and diversity. The system has been fully developed using Java language and it uses other free and standard Java library. The system allows to the user the building of databases of molecules, store information about the molecules, the matching among molecules using different isomorphism paradigms and the similarity/diversity analysis of databases through a wide number of similarity indices.

Hoogsteen sugar pattern of RNA base pairing is studied using density functional theory. Hydrogen bonding patterns of these base pairs are characteristized using NBO analysis and AIM analysis. Correlation between strength of base pairing and the nature of donor-acceptor combinations is also carried out.

Optimal control theory is applied to obtain infrared laser pulses for the selective vibrational excitation of a two mathematical dimensional model of carboxy-myoglobin. Density functional theory is used to obtain the potential energy and dipole moment surfaces of the active site model. The Conjugate gradient method is employed to optimize the cost functional and to obtain the optimized laser pulses. Optimized laser fields are found which give virtually 100% excitation probability to preselected vibrational levels.

Ground state and excitation energies of poly(3-methoxy-thio-phene) (PMT) and poly(thienylene vinylene) (PTV) conjugated polymers are studied by first principles density functional theory (DFT). Two basic approaches of computational chemistry and physics are compared: time dependent DFT (TDDFT) of clusters and

ab initio

pseudopotentials within a standard DFT (PP-DFT) of infinite polymer chains. We demonstrate that series of excitation energies of PMT calculated by TDDFT with increased unit numbers converge well to the real experimentally measured energy gaps. Combination of TDDFT cluster method with PP-DFT approach for infinite chain provides single-gap quasiparticle correction value needed for optical calculations. Infinite chain model is used to calculate optical absorption of PTV.

The workshop focus on computational science aspects of asset/derivatives pricing & financial risk management that relate to business intelligence. It will include but not limited to modeling, numeric computation, algorithmic and complexity issues in arbitrage, asset pricing, future and option pricing, risk management, credit assessment, interest rate determination, insurance, foreign exchange rate forecasting, online auction, cooperative game theory, general equilibrium, information pricing, network band witch pricing, rational expectation, repeated games, etc.

Complex financial instruments are a central concept for the survival of financial enterprises in liberalized markets. The need for fast pricing of more complex and exotic financial products led to the development of new algorithms, and to the parallelization of existing algorithms. In this paper, we present a parallelization scheme for pricing path-dependent interest rate products on bounded trinomial lattices. The basic building block presented in this paper can be used to build more complex pricing schemes. The paper is concluded by a set of numerical results concerning the speedup of the proposed parallelization scheme.

We present a nonlinear structural stock market model which is a nonlinear deterministic process buffeted by dynamic noise. The market is composed of two typical trader types, the rational fundamentalists believing that the price of an asset is determined solely by its fundamental value and the boundedly rational noise traders governed by greed and fear. The interaction among heterogeneous investors determines the dynamics and the statistical properties of the system. We find the model is able to generate time series that exhibit dynamical and statistical properties closely resembling those of the S&P500 index, such as volatility clustering, fat tails (leptokurtosis), autocorrelation in square and absolute return, larger amplitude, crashes and bubbles. We also investigate the nonlinear dependence structure in our data. The results indicate that the GARCH-type model cannot completely account for all nonlinearity in our simulated market, which is thus consistent with the results from real markets. It seems that the nonlinear structural model is more powerful to give a satisfied explanation to market behavior than the traditional stochastic approach.

This work furnished a sharper bound of exponential form for the

L

2

norm of an arbitrary shaped random matrix. Based on the newly elaborated bound, a non-uniform sampling method was developed to succinctly approximate a matrix with a sparse binary one and hereby to relieve the computation loads in both time and storage. This method is not only pass-efficient but query-efficient also since the whole process can be completed in one pass over the input matrix and the sampling and quantizing are naturally combined in a single step.

Regularized multiple-criteria linear programming (RMCLP) model is a new powerful method for classification in data mining. Taking account of every training instance, RMCLP is sensitive to the outliers. In this paper, we propose a sample selection method to seek the representative points for RMCLP model, just as finding the support vectors to support vector machine (SVM). This sample selection method also can exclude the outliers in training set and reduce the quantity of training samples, which can significantly save costs in business world because labeling training samples is usually expensive and sometimes impossible. Experimental results show our method not only reduces the quality of training instances, but also improves the performance of RMCLP.

This paper presents a generative approach to direction-of-change time series forecasting. Kernel methods are used to estimate densities for the distribution of positive and negative returns, and these distributions are then combined to produce probability estimates for return forecasts. An advantage of the technique is that it involves very few parameters compared to regression-based approaches, the only free parameters being those that control the shape of the windowing kernel. A special form is proposed for the kernel covariance matrix. This allows recent data more influence than less recent data in determining the densities, and is important in preventing overfitting. The technique is applied to predicting the direction of change on the Australian All Ordinaries Index over a 15 year out-of-sample period.

Optimization-based classification approaches have well been used for decision making problems, such as classification in data mining. It considers that the contributions from all the attributes for the classification model equals to the joint individual contribution from each attribute. However, the impact from the interactions among attributes is ignored because of linearly or equally aggregation of attributes. Thus, we introduce the generalized Choquet integral with respect to the non-additive measure as the attributes aggregation tool to the optimization-based approaches in classification problem. Also, the boundary for classification is optimized in our proposed model compared with previous optimization-based models. The experimental result of two real life data sets shows the significant improvement of using the non-additive measure in data mining.

This work applied data analysis method of attribute reduction to the credit risk evaluation indicators selection of the emerging technical firms (ETF), and constructed the ETF’s credit risk evaluation indicators system. Furthermore, the utilization the distinct matrix method was carried out the confirmation to the reduction result. Finally the 7 ETF’s data was used to carry out the analysis in the western of China by attribute reduction and their core attribute was obtained.

The mobile market is becoming more competitive. Mobile operators having been focusing on the market share of high quality customers. In this paper, we propose a new method to help mobile operator to estimate the share in high quality customers market based on the available data, inter-network calling detail records. The core of our method is a discretization algorithm which adopts the Gini criterion as discretization measure and is supervised, global and static. In order to evaluate the model, we use the real life data come from one mobile operator in China mainland. The results prove that our method is effective. And also our method is simple and easy to be incorporated into operation support system to predict periodically

It is well known that data mining is a process of discovering unknown, hidden information from a large amount of data, extracting valuable information, and using the information to make important business decisions. And data mining has been developed into a new information technology, including regression, decision tree, neural network, fuzzy set, rough set, support vector machine and so on. This paper puts forward a rough set-based multiple criteria linear programming (RS-MCLP) approach for solving classification problems in data mining. Firstly, we describe the basic theory and models of rough set and multiple criteria linear programming (MCLP) and analyse their characteristics and advantages in practical applications. Secondly, detailed analysis about their deficiencies are provided respectively. However, because of the existing mutual complementarities between them, we put forward and build the RS-MCLP methods and models which sufficiently integrate their virtues and overcome the adverse factors simultaneously. In addition, we also develop and implement these algorithm and models in SAS and Windows platform. Finally, many experiments show that RS-MCLP approach is prior to single MCLP model and other traditional classification methods in data mining.

Traditional approach to predict large-scale sequential curves is to build model separately according to every curve, which causes heavy and complicated modeling workload inevitably. A new method is proposed in this paper to solve this problem. By reducing model types of curves, clustering curves and modeling by clusters, the new method simplifies modeling work to a large extent and reserves original information as possible in the meantime. This paper specifies the theory and algorithm, and applies it to predict GDP curves of multi-region, which confirms practicability and validity of the presented approach.

As the real estate market develops rapidly and is increasingly securitized, it has become an important investment asset in the portfolio design. Thus the measurement of its market risk exposure has attracted attentions from academics and industries due to its peculiar behavior and unique characteristics such as heteroscedasticity and multi scale heterogeneity in its risk and noise evolution etc. This paper proposes the wavelet denoising ARMA-GARCH approach for measuring the market risk level in the real estate sector. The multi scale heterogeneous noise level is determined in the level dependent manner in wavelet analysis. The autocorrelation and heteroscedasticity characteristics for both data and noises are modeled in the ARMA-GARCH framework. Experiment results in Chinese real estate market suggest that the proposed methodology achieves the superior performance by improving the reliability of VaR estimated upon those from traditional ARMA-GARCH approach.

In this paper we explore the impact of trading taxes (commissions) on day-of-the-week effect in the Lithuanian Stock market. We applied the computational model for processing trading activities only on the particular days of the week. The suggested algorithm of trading shares not only reveals presence of the day-of-the-week anomaly, but allows comparing it to the influence of the trading taxes by estimating the final return of the selected shares. As the taxes of each transaction depend on the investment sum, therefore the suggested algorithm had to optimize the number of operations for ensuring the biggest gain. The research revealed significance of intra-week stock return seasonality for majority of shares (17 out of total 24). The advantages of the suggested method include its ability to better specify the shares for performing intra-week seasonality-based transactions, even though embracing of the trading commissions reduces visibility of the effect.

Information systems have to respond well to the changing business environment. Thus, they must have architecture which withstands the change. To design such systems, business process modeling is effective, however, the models include often abstractness and arbitrariness. Therefore, there have been efforts that validate rigorousness of the models. They have defined semantics of the models and applied various logics and formal methods to verification of the rigorousness. This paper focuses on formal verification of the models and surveys the efforts. We also discuss the prospect of the solutions. The establishment of the verification will be surely helpful toward solving the problems on business process reengineering, business process management, service-oriented architecture, and so on.

Networks representing complex dynamical systems often have a multiscale structure and additional properties of nodes and links, which may vary in time. We briefly summarize current trends in the statistical physics research of networks structure and dynamics and its applications to physical, biological, and social systems discussed in the Workshop.

Topological properties of metabolic networks may reflect systematic differences between evolutionary distinct groups of organisms. Indeed, the mean shortest path length between metabolites is, on average, longer in eukaryotes than in bacteria. We show that not only the averages of groups differ, but the organisms can be successfully clustered, based on network properties, into categories corresponding to taxonomic groups. We use the fact that in metabolic networks of different organisms, correspondence between vertices is available. We compare our approach with several graph indices employed previously to analyse metabolic networks, and show that they fail at achieving level of clustering similar to ours. Finally, we show that the phylogenetic tree constructed using network-based approach agrees in most cases with gene-based phylogeny.

We study an dynamical model on complex networks. The model is based on a multi agent modeling which is aimed for representing ‘Network Effect’ in a market of personal communication services. By series of numerical simulations using various models of complex networks, we found two classes in resulting dynamics which are dependent on underlying network structures as well as parameter settings. We also apply the model on a real social network data, and discuss the simulation results in relation with the network models.

We consider a three dimensional spatial network, where

N

nodes are randomly distributed within a cube

L

×

L

×

L

. Each two nodes are connected if their mutual distance does not excess a given cutoff

a

. We analyse numerically the probability distribution of the critical density

$\rho_c=N(a_c/L)^3$

, where one or more nodes become separated;

ρ

c

is found to increase with

N

as

N

0.105

, where

N

is between 20 and 300. The results can be useful for a design of protocols to control sets of wearable sensors.

Real-data networks often appear to have strong modularity, or network-of-networks structure, in which subgraphs of various size and consistency occur. Finding the respective subgraph structure is of great importance, in particular for understanding the dynamics on these networks. Here we study modular networks using generalized method of maximum likelihood. We first demonstrate how the method works on computer-generated networks with the subgraphs of controlled connection strengths and clustering. We then implement the algorithm which is based on weights of links and show it’s efficiency in finding weighted subgraphs on fully connected graph and on real-data network of yeast.

The spectral properties of the adjacency (connectivity) and distance matrix for various types of networks: exponential, scale-free (Albert–Barabási) and classical random ones (Erdős–Rényi) are evaluated. The graph spectra for dense graph in the Erdős–Rényi model are derived analytically.

Topological, algebraic and combinatorial properties of simplicial complexes which are constructed from networks (graphs) are examined from the statistical point of view. We show that basic statistical features of scale free networks are preserved by topological invariants of simplicial complexes and similarly statistical properties pertaining to topological invariants of other types of networks are preserved as well. Implications and advantages of such an approach to various research areas involving network concepts are discussed.

In this paper we investigate the users’ recommendation networks based on the large data set from the Internet Movie Database. We study networks based on two types of inputs: first (monopartite) generated directly from the recommendation lists on the website, and second (bipartite) generated through the users’ habits. Using a threshold number of votes per movie to filter the data, we actually introduce a control parameter, and then by tuning this parameter we study its effect on the network structure. From the detailed analysis of both networks we find that certain robust topological features occur independently from the value of the control parameter. We also present a comparison of the network clustering and shortest paths on the graphs with a randomized network model based on the same data.

The dynamics of coupled 2D chaotic maps with time-delay on a scalefree-tree is studied, with different types of the collective behaviors already been reported for various values of coupling strength [1]. In this work we focus on the dynamics’ time-evolution at the coupling strength of the stability threshold and examine the properties of the

regularization

process. The time-scales involved in the appearance of the

regular state

and the

periodic state

are determined. We find unexpected regularity in the the system’s final steady state: all the period values turn out to be integer multiples of one among given numbers. Moreover, the period value distribution follows a power-law with a slope of -2.24.

Graph visualization represents an important computational tool in analysis of complex networks. Recently, variety of network structures in complex dynamical systems have been found which require appropriately adjusted visualization algorithms. We are testing quantitatively performance of two visualization algorithms based on energy minimization principle on variety of complex networks from cell-aggregated planar graphs to highly clustered scale-free networks. We found that fairly large structures with high clustering can be efficiently visualized with spring energy model with truncated interaction.

This paper presents the introduction to Geocomputation workshop in ICCS2008. The Workshop on Geocomputation continues with the ICCS conferences held in Amsterdam (2002), St. Petersburg (2003), Krakow (2004), Atlanta (2005), Reading (2006), and Beijing (2007).

Providing geo-spatial data services (GDS) and processing functionality services (PFS) are the key issues in spatial information grid (SIG). Especially, it’s crucial for SIG to offer PFS related to Geographic Information Science (GIS), instead of just focused on Remote Sensing (RS) field. Furthermore, implementing high-performance GIServices is the main task of SIG to offer PFS for GIS. Lacking of high-performance GIServices mainly resulted from the limitations of architecture as well as the complexity for services implementation and encapsulation. Based on existing SIG platform, we propose the new architecture of SIG, upon which the constituted GIS nodes can provide GIServices. Within the improved architecture, some parallel GRASS GIS algorithms programs, which are built by different parallelization patterns and can run in cluster with better efficiency, are encapsulated to high-performance GIServices guiding by certain generic mode. Lastly, the analyses of the test demonstrate that the approach can reach our aims.

The present paper deals with the numerical simulation of threshold-crossing problem allowing us to assess the probability that a random field of contamination does not exceed a fixed level in a certain two-dimensional (2-D) spatial domain. A real-valued, homogeneous random field described by the mean value and the covariance (differentiable) function is assumed as the basic theoretical model of the contamination field. In the numerical simulation, a suitable discrete model defined on a regular or irregular grid has been developed and tested by the conditional simulation method. The practical example concerns a case study of heavy metals concentration in soil of the northern part of Poland. The results of the study indicate that theoretical modelling of the level crossings in 2-D random fields with the continuous parameter shows a good agreement with the numerical simulations of the fields with the discretised parameter.

Prototyping urban road network routes improves the accessibility of road layouts and enhances people’s use of road networks by providing a framework for the analysis and evaluation of routes based on multiple criteria, such as spatial quality, transportation cost and aesthetics. However features identified in this way do not, often, incorporate information about current road condition in order to proactively provide real-time contextual information to users. Context-aware computing has the potential to provide useful information about current road condition by leveraging on contextual information of people, places and things. A prototype, which uses context information and a range of design and user-related criteria to analyse the accessibility of road layouts and plan routes, is developed. A case study is used to validate the prototype.

This paper presents the design, analysis and implementation of InterCondor system. The InterCondor system is an implementation of the concept of InterGrid. It uses Condor as a basic local Grid computing engine. It utilizes a series of Grid services, which including register service, data transfer service, task schedule service, security authentication service and status monitor service, to manage the resources such as remote sensing algorithms, remote sensing data, and computing resource under the management of Condor engine. We aim at integrating Grid Service data management, task schedule, and the computing power of Condor into remote sensing data processing and analysis to reduce the processing time of a huge amount of data and long-processing-time remote sensing task by algorithms issuance, data division, and the utilization of any computing resources unused on Internet.

Spherical Harmonic Transforms (SHTs) which are essentially Fourier transforms on the sphere are critical in global geopotential and related applications. Among the best known strategies for discrete SHTs are Chebychev quadratures and least squares. The numerical evaluation of the Legendre functions are especially challenging for very high degrees and orders which are required for advanced geocomputations. The computational aspects of SHTs and their inverses using both quadrature and least-squares estimation methods are discussed with special emphasis on numerical preconditioning that guarantees reliable results for degrees and orders up to 3800 in REAL*8 or double precision arithmetic. These numerical results of spherical harmonic synthesis and analysis using simulated spectral coefficients are new and especially important for a number of geodetic, geophysical and related applications with ground resolutions approaching 5 km.

The emerging class of urgent geoscience workflows are capable of quickly allocating computational resources for time critical tasks. To date, no urgent computing capabilities for data services exists. Since urgent geoscience and Earth science workflows are typically data intensive, urgent data services are necessary so that these urgent workflows do not bottleneck on inappropriately managed or provisioned resources. In this paper we examine emerging urgent Earth and geoscience workflows, the data services used by these workflows, and our proposed urgent data management framework for managing urgent data services.

The Second Workshop on Teaching Computational Science, within the International Conference on Computational Science, provides a platform for discussing innovations in teaching computational sciences at all levels and contexts of higher education. This editorial provides an introduction to the work presented during the sessions.

In this paper the use of metaheuristics techniques in a parallel computing course is explained. In the practicals of the course different metaheuristics are used in the solution of a mapping problem in which processes are assigned to processors in a heterogeneous environment, with heterogeneity in computation and in the network. The parallelization of the metaheuristics is also considered.

In engineering studies, there is a shift to new teaching methodologies with focus on the student involvement, like project-based learning. Project-based learning courses, however, often relay on a previous course where the technical background to be used in the projects is taught, requiring this way two terms for an area. In this paper, we consider a project-based course on computer network applications, which has been designed to cover both the technical and non-technical content in only one term. In the three years we have been teaching this course, our observation based on questionnaires is that the organization of the course allows to successfully reaching the course objectives. We feel, however, that we do not fully exploit the learning potential the course could have in the first few weeks of the course. We describe how the course is organized and the problems we have identified. We propose a project demonstration tool and describe how our solution improves towards our goals. With the proposed tool, the students should better obtain already in the very first days of the course a clear vision about the projects, allowing fully taking advantage of the opportunities which the course offers. format.

In traditional lecture-driven learning, material to be learned is often transmitted to students by teachers. That is, learning is passive. In active learning, students are much more actively engaged in their own learning while educators take a more guiding role. This approach is thought to promote processing of skills and knowledge to a much deeper level than passive learning. In this paper, a research using supporting materials for active e-learning in computational models and related fields is presented. The contributions of this paper are supporting active tools to improve learning and an evaluation of its use in context.

The paper presents an assessment approach to software development process used within students’ team project. The assessment is based on exemplary Process Assessment Model given in ISO/IEC 15504-5 standard. The results of the assessment suggest the area of improvement in our software development process realization. The history, context and basic assumptions established and proposed for the future improvements in the course are given.

Wofford College instituted one of the first undergraduate programs in computational science, the Emphasis in Computational Science (ECS). Besides programming, data structures, and calculus, ECS students take two computational science courses (Modeling and Simulation for the Sciences, Data and Visualization) and complete a summer internship involving computation in the sciences. Materials written for the modeling and simulation course and developed with funding from National Science Foundation served as a basis the first textbook designed specifically for an introductory course in the computational science and engineering curriculum. The successful ECS has attracted a higher percentage of females than in most computer science curricula. The SIAM Working Group on Undergraduate Computational Science and Engineering Education summarized features of Wofford’s ECS and other computational science programs. Besides its established curriculum, Wofford has incorporated computational science in other courses, such as in a sequence of three microbiology laboratories on modeling the spread of disease.

A new experience for teaching Cryptography to engineering students is shown. The aim is to give them a better understanding of secure and cryptographic algorithms by using Maple software, in a graduate-level course. In this paper we discuss how to structure, define, and implement a web-based course as a part of the traditional classes, according to the convergence of the European Higher Education Project. The proposed course facilitates the use of new Information and Communication Technologies.

Currently, European countries are in the process of rethinking their higher education systems due to harmonization efforts initiated by the so-called Bologna’s declaration. This process of reforms implies not only a re-evaluation of our way of teaching and learning but also a new curricular design in order to achieve the expected goals. In this context, the present paper focuses on the problem of teaching computer graphics, a discipline that is getting increasing importance in the realm of Computational Science. The paper discusses some issues regarding an introductory course on computer graphics taking into account its goals, contents, students’ profile and other factors derived from the new scenario given by the European Space of Higher Education.

Self-directed learning has been one of the main objectives in the education domain. A learning model can drive a self-directed learning if an adequate educational environment is built. We propose an educational environment to encourage the self-directed learning, which is composed of a computer collaborative tool that uses dialogues and the concept of ill structured problems. The knowledge being learned is represented by a network of concepts built by the students through the exchanged messages. The network of concepts expressed the relation between the main concepts of the topic being learned. A coherent network is the tangible proof that the process of self-directed learning has been correctly achieved. Two topics of computer sciences are reported: Object Oriented Programming and Case Based Reasoning. The results have proven that along with the knowledge acquired, self-directed learning contributes directly to the development of skills for solving problems and attitudes of collaborative work.

This article describes an innovative methodology for teaching an undergraduate course on Computational Science, with a particular emphasis in Computational Fluid Dynamics (CFD), and the experiences derived from its implementation. The main activities taking place during this course are: development by students of a training project on a topic in materials science, development of a larger CFD project, and an introduction to a CFD commercial package. Projects are carried out by groups of students and are assigned from a set of different available possibilities. Project development consists in the implementation in code of the corresponding mathematical models and a graphical interface which permits the visualization of the results derived from the numerical resolution of the models. The main innovative aspects of the methodology are the use of Project Based Learning combined with the participation of lecturers from different areas of expertise. Other innovative issues include the opportunity for students to practice skills such as report writing, doing oral presentations, the use of English (a foreign language for them) and the use of Linux as the development environment.