Modeling patterns in temporal data has arisen as an important problem in engineering and science. This has led to the popularity of several dynamic models, in particular the renowned hidden Markov model (HMM) [Rabiner, 1989]. Despite its widespread success in many cases, the standard HMM often fails to model more complex data whose elements are correlated hierarchically or over a long period. Such problems are, however, frequently encountered in practice. Existing efforts to overcome this weakness often address either one of these two aspects separately, mainly due to computational intractability. Motivated by this modeling challenge in many real world problems, in particular, for video surveillance and segmentation, this thesis aims to develop tractable probabilistic models that can jointly model duration and hierarchical information in a unified framework. We believe that jointly exploiting statistical strength from both properties will lead to more accurate and robust models for the needed task. To tackle the modeling aspect, we base our work on an intersection between dynamic graphical models and statistics of lifetime modeling. Realizing that the key bottleneck found in the existing works lies in the choice of the distribution for a state, we have successfully integrated the discrete Coxian distribution [Cox, 1955], a special class of phase-type distributions, into the HMM to form a novel and powerful stochastic model termed as the Coxian Hidden Semi-Markov Model (CxHSMM). We show that this model can still be expressed as a dynamic Bayesian network, and inference and learning can be derived analytically. / Most importantly, it has four superior features over existing semi-Markov modelling: the parameter space is compact, computation is fast (almost the same as the HMM), close-formed estimation can be derived, and the Coxian is flexible enough to approximate a large class of distributions. Next, we exploit hierarchical decomposition in the data by borrowing analogy from the hierarchical hidden Markov model in [Fine et al., 1998, Bui et al., 2004] and introduce a new type of shallow structured graphical model that combines both duration and hierarchical modelling into a unified framework, termed the Coxian Switching Hidden Semi-Markov Models (CxSHSMM). The top layer is a Markov sequence of switching variables, while the bottom layer is a sequence of concatenated CxHSMMs whose parameters are determined by the switching variable at the top. Again, we provide a thorough analysis along with inference and learning machinery. We also show that semi-Markov models with arbitrary depth structure can easily be developed. In all cases we further address two practical issues: missing observations to unstable tracking and the use of partially labelled data to improve training accuracy. Motivated by real-world problems, our application contribution is a framework to recognize complex activities of daily livings (ADLs) and detect anomalies to provide better intelligent caring services for the elderly. / Coarser activities with self duration distributions are represented using the CxHSMM. Complex activities are made of a sequence of coarser activities and represented at the top level in the CxSHSMM. Intensive experiments are conducted to evaluate our solutions against existing methods. In many cases, the superiority of the joint modeling and the Coxian parameterization over traditional methods is confirmed. The robustness of our proposed models is further demonstrated in a series of more challenging experiments, in which the tracking is often lost and activities considerably overlap. Our final contribution is an application of the switching Coxian model to segment education-oriented videos into coherent topical units. Our results again demonstrate such segmentation processes can benefit greatly from the joint modeling of duration and hierarchy.

<p><p>Det existerar i dagsläget många färdigutvecklade mjukvaror för att underlätta olika organisationers processer.  Dessa mjukvaror täcker ofta stor del av den funktionalitet som organisationen behöver – till ett förhållandevis lågt pris.</p><p>Priset för att skräddarsy mjukvara, i motsats till färdigutvecklad mjukvara, efter sin egen organisation kan, speciellt när företaget är litet, vara väldigt högt i förhållande till lönsamheten. Då alla organisationer är olika, och således har olika funktionalitetskrav kan flera system behövas för att täcka dem. Överlappningen som sker mellan systemen ger plats för redundant data, som utan noggrann kontroll kan bli felaktig data.</p><p>En lösning är att integrera mjukvaror med varandra för att skapa bryggor emellan och således ta bort risken för felaktig data. Detta är ursprungspunkten för denna undersökning. Bristen på information om hur en sådan systemutveckling bör ske är motivet för denna utredning. Utredningens syfte är att täcka detta kunskapsbehov.</p><p>Genom att använda aktionsforskning, som förespråkar nära samarbete mellan praktiker och forskare, har ett organisationsspecifikt problem studerats. Artefakter från vissa delar av systemutvecklingsmetoden Rational Unified Process(RUP) har utvecklats för att ge organisationen ett praktiskt bidrag, samtidigt som ett akademiskt bidrag i form av riktlinjer har utvecklats. Eftersom avgränsning till vissa delar av RUP är gjord utvecklades inte ett fullt system utan artefakter för att underlätta själva utvecklingen – en slags förstudie.</p><p>Riktlinjerna, som speglar det praktiska bidraget, visar på ett ökat behov av att beskriva och illustrera arkitekturen hos det system som ska utvecklas. Men även ytterligare kunskap om de system som skall integreras bör utvecklas, detta innebär ett behov av närmare samarbete med organisationen.</p><p>RUP kan kompletteras med ytterligare metoddelar för att täcka speciella behov som projektet kan ha – detta inkluderar metoddelar för att hantera integrationsprojekt. Det är denna undersöknings slutsats att sådan komplettering inte är nödvändig i mindre projekt då RUP redan beskurits ordentligt under situationsanpassningen. Artefakter för att beskriva arkitektuella aspekter av projektet finns i RUP och inkluderingen av dessa bör täcka de behov ett mindre projekt har.</p><p>Undersökningen har även studerat hur integrationsfokuset påverkat valda delar av RUP. Resultatet av denna frågeställning är ytterligare modell-element i några artefakter samt ett antal ytterligare artefakter för att beskriva arkitektur.</p></p>

No description available.

Thesis (Ph.D.)--University of Delaware, 2007. / Principal faculty advisor: Qaisar Shafi, Dept. of Physics & Astronomy. Includes bibliographical references.

One important aspect of the quality of programming languages is the effort required by a programmer to understand code written in the language. A historical case where this issue was at the forefront was in the debate between the proponents of high-level languages (HLL) and Assembly languages, where the main argument for HLLs were that they were easier for people to understand. Being one out of a series of articles arguing for a unified theory for software engineering, this article proposes the use of a specific theoretical model from the discipline of cognitive psychology as a tool for predicting language comprehension effort. Describing human problem solving faculties, the ACT-R model [Anderson and Lebiere 1998] predicts that the effort of understanding a program written in C is only 36,5% of the effort of understanding a comparable program written in Assembly. In order to validate the theory, an experiment was performed where a number of engineering students were exposed to tasks of program comprehension. This empirical assessment demonstrated that the effort of understanding a program written in C is 32,5% of the effort of understanding a comparable program written in Assembly. Comparing the results of the theoretical predictions and the empirical assessments of program comprehension effort, we find that the theoretical model performs surprisingly well. The prediction error for the execution of an Assembly program was 5,1% while the error for C was 6,8%. The prediction error for the ratio between the two program languages amounted to 12,6%. / <p>QC 20130618</p>

The goal of this thesis is to develop an easily modifiable sensor system. To achieve this goal SensorML (an XML based sensor language) is combined with Java Beans (a component model language). An important part of SensorML is its process model. Each sensor in the real world is depicted in SensorML by a process model, whereas the connections between the sensors are shown by a process chain. This thesis presents a translator that reads these documents and converts them to Java Beans. Through testing the Translator is proved more efficient than the convenient Object Oriented approach. / UOIT

This study examines the effect that divided or unified government, in the United States of America, has on the delegation of trade authority to the President. Using a qualitative analysis approach, I examine competing views and formulate an independent opinion based on the peoples’ preferences and evaluation of the principles of America’s Constitutionalism. I conclude that overemphasis on the impact of divided government is misleading because trade issues provide the primary mechanism which determines the implementation of America’s trade policies, and the principles of Constitutionalism are valuable guidelines. Blended with the discussion is the awareness of an American ethos which challenges formulation of trade agreements in an era of increased globalization.

No description available.

This thesis describes a compiler optimization targeted for machines with unified, vector-based register sets. This optimization combines register allocation and instruction scheduling. It examines places where the code performs computations on scalar variables. The goal is to identify instances where the same operation is performed. For example, a program might calculate ¡§base+offset¡¨ and then calculate ¡§i+j¡¨. Even though these computations are unrelated, yet they use the same operator; if ¡§base¡¨ and ¡§i¡¨ are packed into one vector register, while ¡§offset¡¨ and ¡§j¡¨ are packed into another, then these two computations can be performed simultaneously through the vectors¡¦ parallel addition operation. This would reduce the execution time of the compiled code. Although other researchers have considered similar packing methods, their work has been limited by the hardware that they were studying. Such hardware usually imposed high costs for moving data between scalar and vector register banks. This present thesis, however, considers a novel hardware architecture that imposes no such costs. As a consequence, we are able to obtain significant speedups. The architecture that we consider is a Graphics Processing Unit (GPU) for embedded systems that is under development at this university. This GPU has a single register set for integers, float, and vectors.

As a result of Internet popularly, the World Wide Web, a killer application which improves the global network, is used much more frequently. Because World Wide Web is browsed through GUI (Graphic User Interface) which allows end-user to interact and operate with internet websites easily, more and more websites are created and operating. For this moment, the web based GUI proved its advantages that it can be applied not only to general website but also to electronic commerce, enterprise information system, and system resources administration. Today, the processing architecture evolves from early centralized processing architecture to recently distributed processing architecture which is composed of several servers and in which each server is responsible for a different system service. But because each system platform runs on its own unique operating system, system administrators have difficulties operating it. Due to increased numbers of servers, a well-defined systematic management architecture and remote management is necessary. This thesis contains the designs of web based system administration architecture through the designs of united library interface to increase the convenience of expansion on library. In this paper, a prototype system created during the research development will serve as theory verification and as an example for developers of future system management platforms.