Browsing Doctoral Dissertations by Title
-
Bojović, R. Dejan (Kragujevac, Serbia , 1999)[more][less]
Abstract: The field of research in this dissertation is consideration of convergence of finite differnce method for parabolic problems with generalized solutions. The research in this dissertation is connected with the following subjects: Numerical Analysis and Partial Differential Equations. This dissertation, beside Preface and References with 56 items, consists of five chapters: 1. Introductory Topics; 2. Parabolic Problems with Variable Operator: Convergence in W(2,1)-norm; 3. Parabolic Problems with Variable Operator: Convergence in W(1,1/2)-norm; 4. Convergence in L-2 norm; 5. Application of Interpolatyion theory In Chapter 1 a brief review of the Sobolev spaces, anisotropic Sobolev spaces, multipliers in Sobolev spaces, interpolation theory of Banach spaces and existence of generalized solution of parabolic problems are presented. Initial-boundary-value problems with variable (time-dependent) operator are considered in Chapters 2 and 3. In Chapter 2 is proved convergence of finite difference scheme in discrete W(2,1) Sobolev norm. Convergence in W(1,1/2) norm is proved in Chapter 3. In Chapter 4, parabolic problem with variable coefficients is considered and convergence in L-2 norm is proved. Finally, in Chapter 5 , interpolation theory is applied to the convergence analysis. URI: http://hdl.handle.net/123456789/1915 Files in this item: 1
doktorska disertacija Scan reduce.pdf ( 1.523Mb ) -
Hodžić, Sandra (Beograd , 2016)[more][less]
Abstract: In recent years there has been increasing interest in modeling the physical and chemical processes with equations involving fractional derivatives and integrals. One of such equations is the subdi usion equation which is obtained from the di usion equation by replacing the classical rst order time derivative by a fractional derivative of order with 0 < < 1: The subject of this dissertation is the initial-boundary value problem for the subdi usion equation and its approximation by nite di erences. At the beginning, the one-dimensional equation is observed. The existence and the uniqueness of weak solution is proved. The stability and the convergence rate estimates for implicite and the weighted scheme are obtained. The main focus is on two-dimensional subdi usion problem with Laplace operator as well as problem with general second-order partial di erential operator. It is assumed that the coe cients of the di erential operator satisfy standard ellipticity conditions that guarantees existence of solution in appropriate spaces of Sobolev type. In that case, apart from above mensoned, we constructed the additive and the factorized di erence schemes. We investigated their stability and convergence rate depending on the smoothness of the input data and of generalized solution. URI: http://hdl.handle.net/123456789/4455 Files in this item: 1
Disertacija_Sandra_Hodzic.pdf ( 913.2Kb ) -
Bejtullahu, Rasim (Pristina , 1976)[more][less]
-
Delić, Aleksandra (Beograd , 2016)[more][less]
Abstract: The time fractional di usion-wave equation can be obtained from the classical diffusion or wave equation by replacing the rst or second order time derivative, respectively, by a fractional derivative of order 0 < 2. In particular, depending on the value of the parameter , we distinguish subdi usion (0 < < 1), normal di usion ( = 1), superdi usion (1 < < 2) and ballistic motion ( = 2). Fractional derivatives are non-local operators, which makes it di cult to construct e cient numerical method. The subject of this dissertation is the time fractional di usion-wave equation with coe cient which contains a singular distribution, primarily Dirac distribution, and its approximation by nite di erences. Initial-boundary value problems of this type are usually called interface problems. Solutions of such problems have discontinuities or non-smoothness across the interface, i.e. on support of Dirac distribution, making it di cult to establish convergence of the nite di erence schemes using the classical Taylor's expansion. The existence of generalized solutions of this initial-boundary value problem has been proved. Some nite di erence schemes approximating the problem are proposed and their stability and estimates for the rate of convergence compatible with the smoothness of the solution are obtained. The theoretical results are con rmed by numerical examples. URI: http://hdl.handle.net/123456789/4337 Files in this item: 1
ADelicDisertacija.pdf ( 1.356Mb ) -
Korać, Vanja (Beograd , 2014)[more][less]
Abstract: Digitalna forenzika je multidisciplinarna nauka koja podrazumeva spoj razlicitih nauĉnih disciplina (raĉunarske nauke, pravo, kriminologija) sa brojnim izazovima u uslovima masovnog generisanja digitalnih podataka (Big Data), virtuelizacije klijentske i serverske strane (Cloud Computng), neusaglašenosti standardizacionih tela i opšteg nedostatka brojnih standarda i eksperata u svim disciplinama. Kako se digitalna forenzika odnosi na sve digitalne uraĊaje, uţa nauĉna oblast uklјuĉuje brojne aplikacije digitalne forenzike, kao što su raĉunarska forenzika, forenzika mobilnih ureĊaja, forenzika na sistemima savremenih automobila, senzorskih mreţa itd. U ovom radu je analizirana i primenjena uţa nauĉna oblast raĉunarske forenzike. Opisana je digitalna forenzika raĉunarskih sistema baziranih na Windows i Linux platformi, sa fokusom na odreĊena mesta u implementiranom sistemu proaktivne digitalne forenzike koja mogu ukazati na forenziĉki relevantne dogaĊaje kritiĉne za bezbednost sistema. Opisane su brojne metodologije, tehnologije i tehnike istrage visokotehnološkog kriminala. Proces prikuplјanja podataka i digitalne forenziĉke analize „uţivo―, detalјno je razmatran. Izvršena je kratka revizija karakteristika i tipiĉno zahtevanih funkcionalnosti softverskih forenziĉkih alata, za inicijalni odgovor i oporavak podataka i particija magnetnih diskova. Opisani su i najvaţniji digitalni forenziĉki kompleti alata i njihove osnovne funkcionalnosti. U radu se istiĉu i najznaĉajniji elementi kojima treba posvetiti posebnu paţnju prilikom digitalne forenziĉke analize u virtuelnom okruţenju. TakoĊe su objašnjeni i najvaţniji segmenti samog virtuelnog okruţenja i naĉin na koji oni mogu biti znaĉajni alati, za postupak digitalne forenziĉke analize. U poslednjem delu ovog rada, fokus je usmeren na ranjivosti Windows i Linux platformi sa prikazanim naĉinima zlonamernog proboja sistema. Opisane su opšte ranjivosti i specifiĉne ranjivosti koje se odnose samo na Windows, odnosno samo na Linux platforme. TakoĊe, navedeni su i najĉešći naĉini zlonamernog iskorišćavanja sistema. Ranjivosti raĉunarskih sistema i mreţa mogu se odnositi na programe, hardver, konfiguraciju i lјude. Isklјuĉujući lјude kao najznaĉajniji i istovremeno najkritiĉniji faktor u zaštiti informacija, programske ranjivosti se tipiĉno koriste za online direktne napade, ili napade malicioznim programima. Otkrivanje i otklanjanje ranjivosti sistemskih programa je jedan od glavnih cilјeva digitalne forenzike. Pored skuplјanja forenziĉki relevantnih digitalnih podataka i izgradnje ĉvrstih digitalnih dokaza o kompjuterskom incidentu ili kriminalu za potrebe pravosudnog sistema, cilј digitalne forenziĉke analize je da se iskorišćene ranjivosti trajno otklone i da se incident/protivpravna aktivnost takve vrste više nikada ne ponovi. U tom smislu je doprinos ovog rada veoma znaĉajan. Praktiĉan primer ispitivanja ranjivosti servisa na Windows i Linux platformama obuhvatio je 80 operativnih sistema. Od tog broja, 51 se odnosi na Windows operativne sisteme, a 29 na Linux operativne sisteme. Dobijeni rezultati su rezultat dvogodišnjeg istraţivanja, jer je ispitivanje sistema vršeno u 2011. i 2013. godini. Kroz skeniranje i prikaz ranjivosti difoltno instaliranih Windows i Linux sistema preventivno se otkrivaju ranjivosti koje potencijalno mogu biti iskorišćene od strane bezbednosnih pretnji (maliciozni programi ili zlonamerni napadaĉi) i time ugroziti raĉunarske sisteme i informacije. Proaktivnim otklanjanjem ovih ranjivosti realizuje se preventivna zaštita. Uspostavlјanjem sistema proaktivne forenzike, obezbeĊuje se logovanje forenziĉki relevantnih dogaĊaja, tj. tragova pokušaja napada u realnom vremenu, ĉime se bitno olakšava forenziĉka istraga u sluĉaju incidenta ili protivpravne aktivnosti. URI: http://hdl.handle.net/123456789/3869 Files in this item: 1
doktorat_Vanja_Korac.pdf ( 9.093Mb ) -
Ilić, Dragana (Belgrade , 2008)[more][less]
-
Zougdani, Hassan (Belgrade , 1984)[more][less]
URI: http://hdl.handle.net/123456789/148 Files in this item: 1
phdHassanKhalifaZougdani.pdf ( 4.157Mb ) -
Đurić, Slavko (Kragujevac , 1968)[more][less]
-
Ćirić, Dušan (Belgrade , 1981)[more][less]
-
Zejnullahu, Ramadan (Priština)[more][less]
-
Pevac, Irena (Belgrade)[more][less]
-
Ćiprijanović, Aleksandra (, 2016)[more][less]
Abstract: Studying cosmic rays is very important for better understanding of high energy physical processes, since particles accelerated in space can reach energies far above what we can produce in accelerators on Earth, at the moment. Processes that produce cosmic rays are still unexplained to some extent, and models that have been proposed are awaiting confirmation. Presence of cosmic rays can be measured by detecting different products of interactions of these high energy particles with the interstellar medium through which they propagate. This thesis deals with the accretion shock as sources of cosmic rays in clusters of galaxies, as well as cosmic rays accelerated in supernova remnants inside galaxies. No matter which of these mechanisms is being considered, cosmic rays will collide with atoms and ions in the interstellar medium, and produce, among other things, gamma rays, neutrinos, as well as light elements, of which we will discuss lithium. In the thesis we primarily develop models that describe gamma rays produced by cosmic rays accelerated in shocks that can appear in different processes. We first examine accretion of new gas onto already virialized structures (for example in galaxy clusters). For the first time, we include the change of gamma-ray pro- duction with time, through the history of the universe, that reflects the evolution of accretion shocks which appear during large scale structure formation. Therefore, the models developed in this thesis describe the gamma rays from large scale struc- tures more realistically, compared to models which have previously been developed and which use single redshift approximation for the gamma-ray origin. Models are used to derive the gamma-ray flux of all unresolved galaxy clusters. These mode- led gamma rays are then compared to the isotropic diffuse gamma-ray background, measured by telescope Fermi-LAT. This leads to the conclusion that these cosmic rays have non-negligible contribution to the isotropic diffuse gamma-ray background (depending on the normalization, they can even explain the whole isotropic diffuse gamma-ray background) and that this population of cosmic rays has to be taken into consideration in addition to other components that are thought to be major contributors, like for example, unresolved normal galaxies or blazars. In the thesis, models of gamma-ray production in accretion shocks are also com- pared to observations of high-energy neutrinos detected by IceCube detector. Neu- trinos are used to normalize gamma-ray models, from which we conclude that if the accretion shocks are predominantly strong, neutrino background is more limiting to the possible gamma-ray emissivity of these objects, compared to the gamma-ray background we first used. Study of neutrinos as products of cosmic-ray interactions is very important, since neutrinos interact weakly with other particles, and therefore keep all of the information about the time they were produced and about cosmic rays that produced them. One part of the thesis deals with the production of cosmic rays in supernova remnants, in particular, the case of the Small Magellanic Cloud, which was detected in gamma rays. In this galaxy we also have the first measurements of the lithium abundances in the interstellar gas outside of the Milky Way. Since gamma rays and lithium are produced through interactions of cosmic rays with the interstellar medium, their same origin can be used to estimate the production of lithium and gamma rays by any cosmic-ray population. We show that galactic cosmic rays, which are considered to be dominant population of cosmic rays in the Small Magellanic Cloud, can only explain a very small part of the observed abundance of lithium, if we assume that the entire present gamma-ray emissivity that we observe also originates from the interaction of galactic cosmic rays with gas within the galaxy. This conclusion is interesting, because it leads to the possible existence of other sources of lithium in the Small Magellanic Cloud. Also, using the fact that gamma rays and lithium share the same origin, we estimate how much can irregular dwarf galaxies contribute to the diffuse gamma-ray background. Study of several different products of cosmic-ray interactions with the interstellar medium (gamma rays, neutrinos and lithium) on smaller scales (within the galaxy), as well as on the largest scales (galaxy clusters), showed that in addition to the galactic cosmic rays accelerated in supernova remnants, other still hypothetical co- smic rays (produced for example during accretion of gas on largest scales, or tidal interactions of galaxies) can have a non-negligible contribution to the measurements. URI: http://hdl.handle.net/123456789/4758 Files in this item: 1
AleksandraCiprijanovic.pdf ( 29.40Mb ) -
Bertolino, Milorad (Belgrade)[more][less]
-
Dajović, Vojin (Belgrade)[more][less]
-
Vujičić Stanković, Staša (Beograd , 2016)[more][less]
Abstract: The basic goal of this doctoral thesis is a research into different techniques and models which are applied in information extraction, and providing an informatic support in processing of natural language texts from culinary and gastronomy domain. Information extraction is a subfield of computational linguistics which includes techniques for natural languages processing, in order to find relevant information, define their meaning and establish relations between them. A very special attention is given to ontology based information extraction. It consists of the following: recognition of instances of ontology concepts in non‐structured or semistructured texts written in natural language, reasoning over the identified instances based on the rules defined in the ontology, as well as recognition of instances and their use for instantiating the proper ontology concepts. The main result of thesis reflects in the presentation of a new model for ontology based information extraction. Besides solving tasks of information extraction, the new model includes not only upgrade of existing lexical resources and ontologies, but also creation of the new ones. Its application resulted in development of a system for extraction of information related to the culinary domain, but this new model can be used in other fields as well. Beside this, the food ontology has been developed, Serbian WordNet is extended for another 1.404 synsets from the culinary domain, while electronic dictionary of Serbian is enlarged with 1.248 entries. The significance of the model application comes from the fact that the new and enriched linguistic resources can be used in other systems for natural language processing. The opening chapter of the thesis elaborates the need of providing an informatic model for processing a huge linguistic corpus related to culinary and gastronomy domain, through methodologically precise and solid approach integrating pieces of information on the domain. Also, the formalization of the basic research subject, text in electronic form, has been presented. Further on, the chapter contains a description of the natural languages approximations introduced in order to enable modern information technologies to process texts written in natural languages, and it emphasizes the need to make the characterisation of the text language with corresponding corpus and sublanguage. Further on in the first chapter, the task of information extraction, and the models for informatic processing of non‐structured or semi‐structured texts, used by the computer to interpret the meaning that the author (not necessarily a human) has intended to give while writing the text, are defined. Additionally, this chapter contains the description of the methods used in information extraction field – methods based on rules and methods based on machine learning. Their advantages and shortcomings are listed, so as the reasons why in this thesis are used techniques based on linguistic knowledge. As a conclusion to the introduction chapter, a special attention is given to ontologies, WordNet, and the significance of its usage as ontology. The second chapter contains the presentation of the linguistic resources and tools exploited in this thesis. It describes morphological dictionaries and local grammars used for solving the problem of information extraction from texts written in Serbian. A review of information extraction systems is given subsequently. At the end of the second chapter, the stages in processing of Serbian written texts during the information extraction in the software systems Unitex and GATE are described. The main result of the thesis is presented in the third chapter. It is the model for solving the problem of information extraction by integrating linguistic resources and tools, which includes creation of a text corpus, definition of tasks for information extraction, establishment of finite state models for information extraction, and their application accordingly, iterative enlarging of electronic morphological dictionaries, enrichment and enhancement of WordNet, and creation of new ontologies. Each of these steps is described thoroughly. Even though the model was at first considered as a solution for problems in processing Serbian, it can be equally applied for processing texts written in other languages, with the development of suitable language resources accordingly. The implementation of the above explained steps is described in the fourth chapter, through a system for information extraction from the culinary texts written in Serbian. Then follows the description of a bond in the development and mutual complement of lexical resources through steps in creating domain corpus, identifying culinary lexica, expanding and upgrading of WordNet and electronic morphological dictionaries, and developing of domain ontologies – the food ontology, the approximate measure ontology, and the ontology of ingredients that can be used as mutual replacements in the culinary domain. This system, developed for information extraction, has served for creating an advanced search system which, based on a corpus of culinary texts, generates all possible answers to inquiries made by users. In the frame of this system is implemented a specific method which serves for creation of links between different recipes. This is used in case when the user reviews a text of a recipe and notices that in preparing description features some part which already had appeared in other recipe, but with additional or different explanation. Another contribution of this thesis is application of developed ontologies in tasks that convert approximate measures into standard measures, and establishment of similarities among the recipes. The similarity of the recipes is defined as similarity of texts which describe process of course preparation in accordance with a specific recipe. The last chapter contains final conclusions and directions for future research. URI: http://hdl.handle.net/123456789/4410 Files in this item: 1
teza_Stasa.pdf ( 10.38Mb ) -
Hajduković, Dimitrije (Belgrade)[more][less]
URI: http://hdl.handle.net/123456789/38 Files in this item: 1
phdDimitrijeHajdukovic.pdf ( 1.535Mb ) -
Mališić, Jovan (Belgrade)[more][less]
-
Jovalekić, Milica (Beograd , 2022)[more][less]
Abstract: Let M be a maximum and let N be a minimum of the non-negative martingale X1, X2, . . . , Xn. It is well known, that if X1 = 1, then γ(‖M ‖1) ≤ E (Xn log Xn) and γ(‖N ‖1) ≤ E (Xn log Xn) , where γ(x) = x − 1 − log x, for all x > 0. In this thesis, we prove the analogue of this result in the case when 1 < p < ∞, by proving that δp (‖M ‖p p ) ≤ ‖Xn‖p and δp (‖N ‖p p ) ≤ ‖Xn‖p, where δp(x) = ( 1 − 1 p ) x 1 p + 1 p x 1 p −1, for all x > 0. We also obtain a probabilistic proof of the fact min ρ∈D(Qn) ∫ Qn dx1 . . . dxn ρ (x1, . . . , xn)p−1 ∏n j=1 xαj +1 j = n∏ j=1 ( p p − αj − 1 )p , where p > 1, αj < p − 1 for j = 1, . . . , n and D (Qn) is family of all densities on the n-dimensional unit cube Qn = (0, 1)n in Rn. This provides the proof of the multidimensional weighted Hardy inequality. Namely, if f : Rn + → (0, ∞) is a measurable function, p > 1 and αj < p − 1 for j = 1, . . . , n, then ∫ Rn + n∏ j=1 xαj j Hnf (x)p dx ≤ n∏ j=1 ( p p − αj − 1 )p ∫ Rn + n∏ j=1 xαj j f (x)p dx, where Hnf (x) = 1 x1 . . . xn ∫ x1 0 · · · ∫ xn 0 f (t) dt, is a multidimensional Hardy operator, x = (x1, . . . , xn) ∈ Rn +, t = (t1, . . . , tn) and dt = dt1 . . . dtn. Let B(t) be a standard planar Brownian motion and r(θ) be the length of the projection of B[0, 1] on the line generated by the unit vector eθ = (cos θ, sin θ), where 0 ≤ θ ≤ π. We nd the common distribution function F of the random variables r(θ). Namely, we prove that F(x) = 8 ∞∑ n=1 ( 1 x2 + 1 (2n − 1)2π2 ) exp ( − (2n − 1)2π2 2x2 ) , for every x > 0. As immediate consequence, lower bound for the expected diameter of the set B[0, 1], better than known, is obtained. Namely, it is known that Ed ≥ 1.601, where d is the diameter of the set B[0, 1]. In this thesis we show Ed ≥ 1.856. URI: http://hdl.handle.net/123456789/5323 Files in this item: 1
MilicaJovalekicDisertacija.pdf ( 1.796Mb ) -
Zlatanović, Milan (Niš , 2010)[more][less]
-
Milošević, Stefan (Beograd , 2017)[more][less]
Abstract: In this paper we present some norm inequalities for certain elementary operators and inner product type transformers, specially for Schatten norms, if the families of operators generating those transforms consists of arbitrary operators, and Q norms if at least one of those families consists of mutually commuting normal operators. Among others, we present inequalities that are generalizing the inequality p IA AX p IB B 6 X AXB ; from [11, Th. 2.3], for normal contractions and arbitrary unitarily invariant norm, to the case of Schatten norms and arbitrary contractions, as well as Q norms if at least one of the contractions A or B is normal. Also, by applying norm inequalities for operator monotone and operator convex functions, some refined Cauchy - Schwarz operator inequalities, as well as Minkowski and Landau - Gruss norm inequalities for operators are obtained as well. URI: http://hdl.handle.net/123456789/4656 Files in this item: 1
Stefan_Milosevic_Disertacija.pdf ( 2.544Mb )