DSpace Community: 2010
http://hdl.handle.net/10525/1486
The Community's search engineSearch the Channelsearch
http://sci-gems.math.bas.bg/jspui/simple-search
A 3D Model as a Tool for Increasing the Effectiveness of E-Learning
http://hdl.handle.net/10525/1607
Title: A 3D Model as a Tool for Increasing the Effectiveness of E-Learning<br/><br/>Authors: Valcheva, Donika; Todorova, Margarita; Asenov, Oleg<br/><br/>Abstract: The paper proposes a 3D model which could be used as a toolfor increasing the effectiveness of e-learning. It also offers an approach forapplying this 3D model for increasing the effectiveness of e-learning. Thisapproach has methodical value in line with the idea for dynamic adjustmentof the individual learning profile of each student in order to increase thepersonalization level in the e-learning process.Using Inside-Outside Algorithm for Estimation of the Offspring Distribution in Multitype Branching Processes
http://hdl.handle.net/10525/1606
Title: Using Inside-Outside Algorithm for Estimation of the Offspring Distribution in Multitype Branching Processes<br/><br/>Authors: Daskalova, Nina<br/><br/>Abstract: Multitype branching processes (MTBP) model branching structures, where the nodes of the resulting tree are particles of different types.Usually such a process is not observable in the sense of the whole tree, butonly as the “generation” at a given moment in time, which consists of thenumber of particles of every type. This requires an EM-type algorithm toobtain a maximum likelihood (ML) estimate of the parameters of the branching process. Using a version of the inside-outside algorithm for stochasticcontext-free grammars (SCFG), such an estimate could be obtained for theoffspring distribution of the process.On the Arithmetic of Errors
http://hdl.handle.net/10525/1605
Title: On the Arithmetic of Errors<br/><br/>Authors: Markov, Svetoslav; Hayes, Nathan<br/><br/>Abstract: An approximate number is an ordered pair consisting of a (real)number and an error bound, briefly error, which is a (real) non-negativenumber. To compute with approximate numbers the arithmetic operationson errors should be well-known. To model computations with errors oneshould suitably define and study arithmetic operations and order relationsover the set of non-negative numbers. In this work we discuss the algebraicproperties of non-negative numbers starting from familiar properties of realnumbers. We focus on certain operations of errors which seem not to havebeen sufficiently studied algebraically. In this work we restrict ourselves toarithmetic operations for errors related to addition and multiplication byscalars. We pay special attention to subtractability-like properties of errorsand the induced “distance-like” operation. This operation is implicitly usedunder different names in several contemporary fields of applied mathematics(inner subtraction and inner addition in interval analysis, generalizedHukuhara difference in fuzzy set theory, etc.) Here we present some newresults related to algebraic properties of this operation.Solving the Task Assignment Problem with a Variable Neighborhood Search
http://hdl.handle.net/10525/1604
Title: Solving the Task Assignment Problem with a Variable Neighborhood Search<br/><br/>Authors: Kratica, Jozef; Savić, Aleksandar; Filipović, Vladimir; Milanović, Marija<br/><br/>Abstract: In this paper a variable neighborhood search (VNS) approachfor the task assignment problem (TAP) is considered. An appropriate neighborhood scheme along with a shaking operator and local search procedureare constructed specifically for this problem. The computational results arepresented for the instances from the literature, and compared to optimalsolutions obtained by the CPLEX solver and heuristic solutions generatedby the genetic algorithm. It can be seen that the proposed VNS approachreaches all optimal solutions in a quite short amount of computational time.A Bayesian Spatial Mixture Model for FMRI Analysis
http://hdl.handle.net/10525/1603
Title: A Bayesian Spatial Mixture Model for FMRI Analysis<br/><br/>Authors: Geliazkova, Maya<br/><br/>Abstract: We develop, implement and study a new Bayesian spatial mixturemodel (BSMM). The proposed BSMM allows for spatial structure in thebinary activation indicators through a latent thresholded Gaussian Markovrandom field. We develop a Gibbs (MCMC) sampler to perform posteriorinference on the model parameters, which then allows us to assess the posteriorprobabilities of activation for each voxel. One purpose of this articleis to compare the HJ model and the BSMM in terms of receiver operatingcharacteristics (ROC) curves. Also we consider the accuracy of the spatialmixture model and the BSMM for estimation of the size of the activationregion in terms of bias, variance and mean squared error. We perform asimulation study to examine the aforementioned characteristics under a varietyof configurations of spatial mixture model and BSMM both as the sizeof the region changes and as the magnitude of activation changes.An Algorithmic Solution for Management of Related Text Objects with Application in Phytopharmacy
http://hdl.handle.net/10525/1602
Title: An Algorithmic Solution for Management of Related Text Objects with Application in Phytopharmacy<br/><br/>Authors: Dimova, Delyana<br/><br/>Abstract: This paper presents an algorithmic solution for managementof related text objects, in which are integrated algorithms for their extraction from paper or electronic format, for their storage and processing in arelational database. The developed algorithms for data extraction and dataanalysis enable one to find specific features and relations between the textobjects from the database. The algorithmic solution is applied to data fromthe field of phytopharmacy in Bulgaria. It can be used as a tool and methodology for other subject areas where there are complex relationships betweentext objects.Teaching Materials Repository
http://hdl.handle.net/10525/1601
Title: Teaching Materials Repository<br/><br/>Authors: Stanchev, Peter; Nisheva-Pavlova, Maria; Geske, John<br/><br/>Abstract: The paper presents results from the development of a methodology and corresponding software tools for building an academic repository.The repository was filled up with gaming material. The repository architecture and key features of the search engine are discussed. The emphasisfalls on solutions of the large set of problems concerning the development ofproper mechanisms for semantics-based search in a digital repository.Extension of the C-XSC Library with Scalar Products with Selectable Accuracy
http://hdl.handle.net/10525/1600
Title: Extension of the C-XSC Library with Scalar Products with Selectable Accuracy<br/><br/>Authors: Zimmer, Michael; Krämer, Walter; Bohlender, Gerd; Hofschuster, Werner<br/><br/>Abstract: The C++ class library C-XSC for scientific computing has beenextended with the possibility to compute scalar products with selectable accuracy in version 2.3.0. In previous versions, scalar products have alwaysbeen computed exactly with the help of the so-called long accumulator. Additionally, optimized floating point computation of matrix and vector operations using BLAS-routines are added in C-XSC version 2.4.0. In this articlethe algorithms used and their implementations, as well as some potentialpitfalls in the compilation, are described in more detail. Additionally, thetheoretical background of the employed DotK algorithm and the necessarymodifications of the concrete implementation in C-XSC are briefly explained.Run-time tests and numerical examples are presented as well.Quadratic Time Computable Instances of MaxMin and MinMax Area Triangulations of Convex Polygons
http://hdl.handle.net/10525/1599
Title: Quadratic Time Computable Instances of MaxMin and MinMax Area Triangulations of Convex Polygons<br/><br/>Authors: Mirzoev, Tigran; Vassilev, Tzvetalin<br/><br/>Abstract: We consider the problems of finding two optimal triangulations of a convex polygon: MaxMin area and MinMax area. These are thetriangulations that maximize the area of the smallest area triangle in a triangulation, and respectively minimize the area of the largest area trianglein a triangulation, over all possible triangulations. The problem was originally solved by Klincsek by dynamic programming in cubic time [2]. Later,Keil and Vassilev devised an algorithm that runs in O(n^2 log n) time [1]. Inthis paper we describe new geometric findings on the structure of MaxMinand MinMax Area triangulations of convex polygons in two dimensions andtheir algorithmic implications. We improve the algorithm’s running time toquadratic for large classes of convex polygons. We also present experimentalresults on MaxMin area triangulation.Approximating the MaxMin and MinMax Area Triangulations using Angular Constraints
http://hdl.handle.net/10525/1598
Title: Approximating the MaxMin and MinMax Area Triangulations using Angular Constraints<br/><br/>Authors: Mark Keil, J; Vassilev, Tzvetalin<br/><br/>Abstract: We consider sets of points in the two-dimensional Euclideanplane. For a planar point set in general position, i.e. no three points collinear,a triangulation is a maximal set of non-intersecting straight line segmentswith vertices in the given points. These segments, called edges, subdivide theconvex hull of the set into triangular regions called faces or simply triangles.We study two triangulations that optimize the area of the individual triangles:MaxMin and MinMax area triangulation. MaxMin area triangulation is thetriangulation that maximizes the area of the smallest area triangle in thetriangulation over all possible triangulations of the given point set. Similarly,MinMax area triangulation is the one that minimizes the area of the largestarea triangle over all possible triangulations of the point set. For a point setin convex position there are O(n^2 log n) time and O(n^2) space algorithmsthat compute these two optimal area triangulations. No polynomial timealgorithm is known for the general case. In this paper we present an approach<br/><br/>Description: * A preliminary version of this paper was presented at XI Encuentros de Geometr´iaComputacional, Santander, Spain, June 2005.Computer Networks Security Models - A New Approach for Denial-of-Services Attacks Mitigation
http://hdl.handle.net/10525/1597
Title: Computer Networks Security Models - A New Approach for Denial-of-Services Attacks Mitigation<br/><br/>Authors: Tsvetanov, Tsvetomir<br/><br/>Abstract: Computer networks are a critical factor for the performance of amodern company. Managing networks is as important as managing any otheraspect of the company’s performance and security. There are many tools andappliances for monitoring the traffic and analyzing the network flow security.They use different approaches and rely on a variety of characteristics of thenetwork flows. Network researchers are still working on a common approachfor security baselining that might enable early watch alerts. This researchfocuses on the network security models, particularly the Denial-of-Services(DoS) attacks mitigation, based on a network flow analysis using the flowsmeasurements and the theory of Markov models. The content of the papercomprises the essentials of the author’s doctoral thesis.An Improvement to the Achievement of the Griesmer Bound
http://hdl.handle.net/10525/1596
Title: An Improvement to the Achievement of the Griesmer Bound<br/><br/>Authors: Hamada, Noboru; Maruta, Tatsuya<br/><br/>Abstract: We denoted by nq(k, d), the smallest value of n for which an [n, k, d]q code exists for given q, k, d. Since nq(k, d) = gq(k, d) for all d ≥ dk + 1 for q ≥ k ≥ 3, it is a natural question whether the Griesmer bound is attained or not for d = dk , where gq(k, d) = ∑[d/q^i], i=0,...,k-1, dk = (k − 2)q^(k−1) − (k − 1)q^(k−2). It was shown by Dodunekov [2] and Maruta [9], [10] that there is no [gq(k, dk ), k, dk ]q code for q ≥ k, k = 3, 4, 5 and for q ≥ 2k − 3, k ≥ 6. The purpose of this paper is to determine nq(k, d) for d = dk as nq(k, d) = gq(k, d) + 1 for q ≥ k with 3 ≤ k ≤ 8 except for (k, q) = (7, 7), (8, 8), (8, 9).Search for Wieferich Primes through the use of Periodic Binary Strings
http://hdl.handle.net/10525/1595
Title: Search for Wieferich Primes through the use of Periodic Binary Strings<br/><br/>Authors: Dobeš, Jan; Kureš, Miroslav<br/><br/>Abstract: The result of the distributed computing projectWieferich@Homeis presented: the binary periodic numbers of bit pseudo-length j ≤ 3500 obtainedby replication of a bit string of bit pseudo-length k ≤ 24 and increasedby one are Wieferich primes only for the cases of 1092 or 3510.Services for Content Creation and Presentation in an Iconographical Digital Library
http://hdl.handle.net/10525/1594
Title: Services for Content Creation and Presentation in an Iconographical Digital Library<br/><br/>Authors: Pavlov, Radoslav; Paneva-Marinova, Desislava; Goynov, Maxim; Pavlova-Draganova, Lilia<br/><br/>Abstract: Content creation and presentation are key activities in a multimedia digital library (MDL). The proper design and intelligent implementation of these services provide a stable base for overall MDL functionality. This paper presents the framework and the implementation of theseservices in the latest version of the “Virtual Encyclopaedia of BulgarianIconography” multimedia digital library. For the semantic description of theiconographical objects a tree-based annotation template is implemented. Itprovides options for autocompletion, reuse of values, bilingual entering ofdata, automated media watermarking, resizing and conversing. The paperdescribes in detail the algorithm for automated appearance of dependent values for different characteristics of an iconographical object. An algorithm for avoiding duplicate image objects is also included. The service for automatedappearance of new objects in a collection after their entering is included as an important part of the content presentation. The paper also presents theoverall service-based architecture of the library, covering its main servicepanels, repositories and their relationships. The presented vision is basedon a long-term observation of the users’ preferences, cognitive goals, andneeds, aiming to find an optimal functionality solution for the end users.Smartbook: Semantics Inside
http://hdl.handle.net/10525/1593
Title: Smartbook: Semantics Inside<br/><br/>Authors: Koychev, Ivan; Dicheva, Darina; Nikolov, Roumen<br/><br/>Abstract: This paper presents a vision for the future of the e-books whichentails further development of technologies that will facilitate the creationand use of a new generation of "smart" books: e-books that are evolving,highly interactive, customisable, adaptable, intelligent, and furnished witha rich set of collaborative authoring and reading support services. Theproposed set of tools will be integrated into an intelligent framework for collaborative book authoring and experiencing called SmartBook. The paperpromotes the idea that the semantic technologies, intensively developed recently in connection with the Semantic Web initiative, can be incorporatedin the book and become the key factor of making it "smarter".Semantically Enhanced Software Documentation Processes
http://hdl.handle.net/10525/1592
Title: Semantically Enhanced Software Documentation Processes<br/><br/>Authors: Klieber, Werner; Granitzer, Michael; Gaisbauer, Mansuet; Tochtermann, Klaus<br/><br/>Abstract: High-quality software documentation is a substantial issue forunderstanding software systems. Shorter time-to-market software cycles increase the importance of automatism for keeping the documentation up todate. In this paper, we describe the automatic support of the software documentation process using semantic technologies. We introduce a software documentation ontology as an underlying knowledge base. The defined ontology is populated automatically by analysing source code, software documentation and code execution. Through selected results we demonstrate that the use of such semantic systems can support software documentation processes efficiently.