OPUS - Volltextserver Universität Passau
Not a member yet
1021 research outputs found
Sort by
Geographic and Social Space in Latent Factor Models - Four Essays
No full textGeography, social context, time, and cultural mindset are four (out of many) cornerstones of human interaction. When building statistical models, their consideration is vital: They all cause dependency between individual observations, violating assumptions of independence and exchangeability. While this can be problematic and inhibit the unbiased inference of parameters, it can also be a fruitful source of insights and enhance prediction performance.
One class of models that serves to manage or profit from the presence of dependence is the class of latent variable models. This class of models assumes that the presence of non-explicit, unobserved causes of continuous or discrete nature can explain the observed correlations. Latent variable models explicitly take account of dependency, for example, by modeling an unobserved local source of pollution as a continuous spatial variable. Through their widespread use for information fitering, link prediction, and statistical inference, latent variable models have developed an essential impact on our daily life and the way we consume information.
The four articles in this thesis shed light on assumptions, usage, and potential drawbacks of latent variable models in various contexts that involve geographic and interaction data. We model unobserved sources of pollution in geophysical data, explore individual taste and mindsets in cross-cultural contexts, and predict the evolution of social relationships in software development projects. This combination of various perspectives contributes to the interdisciplinary exchange of methodological knowledge on the modeling of dependent data
Adapting Semantic Web Information Retrieval to Multimedia
No full textThe amount of audio, video and image data on the Web is immensely growing, which leads to data management problems based on the hidden character of Multimedia. Therefore the interlinking of semantic concepts and media data with the aim to bridge the gap between the Internet of documents and the Web of Data has become a common practice. However, the value of connecting media to its semantic meta data is limited due to lacking access methods and the absence of an adapted query language specialized for media assets and fragments. This thesis aims to extend the standard query language for the Semantic Web (SPARQL) with media specific concepts and functions. The main contributions of the work are an exhaustive survey on Multimedia query languages of the last 3 decades, the SPARQL extension specification itself and an approach for the efficient evaluation of the new query concepts. Additionally I elaborate and evaluate a meta data based media fragment similarity approach, which provides a basis for further language extensions.Das Wachstum multimedialer Daten wie Audio, Video und Bilder war in den letzten Jahren immens. Das Besondere an dieser Art der Daten ist die versteckte Semantik, die sich nur schwer mit herkömmlichen Information Retrieval Funktionen verbinden lässt und dadurch zu Problemen im Management der Multimedia Daten führt. Konzepte des Semantic Web eignen sich allerdings sehr gut, diese Lücke zu schließen, was sich in vielen Szenarien bereits positiv etabliert hat. Nichtsdestotrotz fehlen mit geeigneten Zugriffsmethoden und einer adaptierten Anfragesprache wichtige Teile, um dieses Konzept der verlinkten Multimedia Daten abzurunden und voll in einem End-to-End Prozess zu verwenden. In dieser Arbeit stelle ich eine Erweiterung der Standard-Anfragesprache im Semantic Web (SPARQL) um multimedia-spezifische Funktionen vor. Der wissenschaftliche Beitrag lässt sich dabei in drei Teile gliedern: Ein umfassendes Survey zu Multimedia Anfragesprachen der letzten 30 Jahre, die Erweiterung von SPARQL inklusive einer geeigneten Methodik zur Anfrageoptimierung, sowie ein Ansatz zur fragment-basierten Ähnlichkeitsberechnung von Bildern mit zugehöriger Evaluierung
Gestufte Standards für die Entwicklung von Kompetenzen in der Lehrerbildung
Get PDFGESTUFTE STANDARDS FÜR DIE ENTWICKLUNG VON KOMPETENZEN IN DER LEHRERBILDUNG (22)
Dimension 1: Gestaltung der Lehrerrolle
Dimension 2: Schule als Lern- und Lebensraum
Dimension 2.1: Schule als Organisation
Dimension 2.2: Schulalltag und Schulleben
Dimension 2.3: Unterrichtsbeobachtung und -evaluation
Dimension 2.4: Entwicklung von Schule
Dimension 3: Unterrichtsplanung, -durchführung und -analyse
Dimension 3.1: Struktur von Unterricht
Dimension 3.2: Lehrformen und -theorien
Dimension 3.3: Lernumgebungen gestalten
Dimension 3.4: Zeitmanagement
Dimension 3.5: Planungsmittel
Dimension 3.6: Medieneinsatz
Dimension 3.7: Sozialformen und Methoden
Dimension 3.8: Umgang mit Heterogenität
Dimension 4: Klassenführung
Dimension 5: Lernprozess- und Lernproduktdiagnostik
Dimension 6: Beratun
Supporting the Discovery of Long-Tail Resources on the Web
No full textA plethora of resources made available via retrieval systems in digital libraries remains untapped in the so called long tail of the Web. These long-tail websites get considerably less visits than major Web hubs.
Zero-effort queries ease the discovery of long-tail resources by proactively retrieving and presenting information based on a user’s context. However, zero-effort queries over existing digital library structures are challenging, since the underlying retrieval system is only accessible via an API. The information need must be expressed by a query, instead of optimizing the ranking between context and resources in the retrieval system directly. We address three research questions that arise from replacing the user information seeking process by zero-effort queries.
Our first question addresses the transformation of a user query to an automatic query, derived from the context. We present means to 1) identify the relevant context on different levels of granularity, 2) derive an information need from the context via keyword extraction and personalization and 3) express this information need in a query scheme that avoids over- or under-specified queries. We address the cold start problem with an approach to bootstrap user profiles from social media, even for passive users.
With the second question, we address the presentation of resources in zero-effort query scenarios, presenting guidelines for presentation interfaces in the browser and a visualization of the triadic relationship between context, query and results. QueryCrumbs, a compact query history visualization supports recalling information found in the past and exploratory search by visualizing qualitative and quantitative query similarity.
Our last question addresses the gap between (simple) keyword queries and the representation of resources by rich and complex meta-data. We investigate and extend feature representation learning techniques centered around the skip-gram model with negative sampling. Finally, we present an approach to learn representations from network and text jointly that can cope with the partial absence of one modality.
Experimental results show close to human performance of our zero-effort query and user profile generation approach and visualizations to be helpful in terms of transparency, efficiency and support for exploratory search. These results indicate that the proposed zero-effort query approach indeed eases the discovery of long-tail resources and the accompanying visualizations further facilitate this process. The joint representation model provides a first step to bridge the gap between query and resource representation and we plan to follow and investigate this route further in the future
Effekte von Immersion und Präsenz auf Lernerfolg in immersiven virtuellen Lernumgebungen zu Themen der Informatik
No full textAbstract concepts and ideas from Computer Science Education can benefit from immersive visualizations that can be provided in virtual environments. This thesis explores the effects of the key characteristics of virtual environments, immersion and presence, on learning outcomes in Educational Virtual Environments for learning Computer Science.
Immersion is a quantifiable description of the technology to immerse the user into the virtual environment; presence describes the subjective feeling of 'being there'. While technological immersion can be seen as a strong predictor for presence, motivational traits, cognition, and the emotional state of the user also influence presence. A possible localization of these technological and person-specific variables in Helmke's pedagogical supply-use framework is introduced as the Educational Framework for Immersive Learning (EFiL). Presence is emphasized as a central criterion influencing immersive learning processes. The EFiL provides an educational understanding of immersive learning as learning activities initiated by a mediated or medially enriched environment that evokes a sense of presence.
The idea of Computer Science Unplugged is pursued by using Virtual Reality technology in order to provide interactive virtual learning experiences that can be accurately displayed, schematizing, substantiating, or metaphorical. For exploring the effects of virtual environment characteristics on learning, the idea of Computer Science Replugged focuses 'hands-on' activities and combines them with immersive technology. By providing a perception of non-mediation, Computer Science Replugged might enable experiences that can contribute additional possibilities to the real activity or enable new activities for teaching Computer Science.
Three game-based Educational Virtual Environments were developed as treatments: 'Bill's Computer Workshop' introduces the components of a computer; 'Fluxi's Cryptic Potions' uses a metaphor to teach asymmetric encryption; 'Pengu's Treasure Hunt' is an immersive visualization of finite state machines. A first study with 23 middle school students was conducted to test the instruments in terms of selectivity, the devices' induced levels of presence, and adequacy of the selected learning objectives. The second study with 78 middle school students playing the environments on different devices (laptop, Mobile Virtual Reality, or head-mounted-display) assessed motivational, cognitive, and emotional factors, as well as presence and learning outcomes.
An overall analysis showed that pre-test performance, presence, and the previous scholastic performance in Maths and German predict the learning outcomes in the virtual environments. Presence could be predicted by the student's positive emotions and by the technological immersion. The level of immersion had no significant effect on learning outcomes. While a good-fitting path analysis model indicated that the assumed relations deriving from the EFiL are largely correct for 'Bill's Computer Workshop' and 'Fluxi's Cryptic Potions', not all results of the overall path analysis were significant for the analyses of the particular environments.
Presence seems to have a small effect on learning outcomes while being influenced by technological and emotional factors. Even though the level of immersion can be used to predict the level of presence, it is not an appropriate predictor for learning outcomes. For future studies, the questionnaires have to be revised as some of them suffered from poor scale reliabilities. While the second study could provide indications that the localization of presence and immersion in an existing educational supply-use framework seems to be appropriate, many factors had to be blanked out.
The thesis contributes to existing research as it adds factors that are crucial for learning processes to the discussion on immersive learning from an educational perspective and assesses these factors in hands-on activities in Educational Virtual Environments for Computer Science Education
Iterative Schedule Optimization for Parallelization in the Polyhedron Model
Get PDFIn high-performance computing, one primary objective is to exploit the performance that the given target hardware can deliver to the fullest. Compilers that have the ability to automatically optimize programs for a specific target hardware can be highly useful in this context. Iterative (or search-based) compilation requires little or no prior knowledge and can adapt more easily to concrete programs and target hardware than static cost models and heuristics. Thereby, iterative compilation helps in situations in which static heuristics do not reflect the combination of input program and target hardware well. Moreover, iterative compilation may enable the derivation of more accurate cost models and heuristics for optimizing compilers. In this context, the polyhedron model is of help as it provides not only a mathematical representation of programs but, more importantly, a uniform representation of complex sequences of program transformations by schedule functions. The latter facilitates the systematic exploration of the set of legal transformations of a given program.
Early approaches to purely iterative schedule optimization in the polyhedron model do not limit their search to schedules that preserve program semantics and, thereby, suffer from the need to explore numbers of illegal schedules. More recent research ensures the legality of program transformations but presumes a sequential rather than a parallel execution of the transformed program. Other approaches do not perform a purely iterative optimization.
We propose an approach to iterative schedule optimization for parallelization and tiling in the polyhedron model. Our approach targets loop programs that profit from data locality optimization and coarse-grained loop parallelization. The schedule search space can be explored either randomly or by means of a genetic algorithm.
To determine a schedule's profitability, we rely primarily on measuring the transformed code's execution time. While benchmarking is accurate, it increases the time and resource consumption of program optimization tremendously and can even make it impractical. We address this limitation by proposing to learn surrogate models from schedules generated and evaluated in previous runs of the iterative optimization and to replace benchmarking by performance prediction to the extent possible.
Our evaluation on the PolyBench 4.1 benchmark set reveals that, in a given setting, iterative schedule optimization yields significantly higher speedups in the execution of the program to be optimized. Surrogate performance models learned from training data that was generated during previous iterative optimizations can reduce the benchmarking effort without strongly impairing the optimization result. A prerequisite for this approach is a sufficient similarity between the training programs and the program to be optimized
Algebraic and Logic Solving Methods for Cryptanalysis
Get PDFAlgebraic solving of polynomial systems and satisfiability of propositional logic formulas are not two completely separate research areas, as it may appear at first sight. In fact, many problems coming from cryptanalysis, such as algebraic fault attacks, can be rephrased as solving a set of Boolean polynomials or as deciding the satisfiability of a propositional logic formula. Thus one can analyze the security of cryptosystems by applying standard solving methods from computer algebra and SAT solving. This doctoral thesis is dedicated to studying solvers that are based on logic and algebra separately as well as integrating them into one such that the combined solvers become more powerful tools for cryptanalysis.
This disseration is divided into three parts. In this first part, we recall some theory and basic techniques for algebraic and logic solving. We focus mainly on DPLL-based SAT solving and techniques that are related to border bases and Gröbner bases. In particular, we describe in detail the Border Basis Algorithm and discuss its specialized version for Boolean polynomials called the Boolean Border Basis Algorithm.
In the second part of the thesis, we deal with connecting solvers based on algebra and logic. The ultimate goal is to combine the strength of different solvers into one. Namely, we fuse the XOR reasoning from algebraic solvers with the light, efficient design of SAT solvers. As a first step in this direction, we design various conversions from sets of clauses to sets of Boolean polynomials, and vice versa, such that solutions and models are preserved via the conversions. In particular, based on a block-building mechanism, we design a new blockwise algorithm for the CNF to ANF conversion which is geared towards producing fewer and lower degree polynomials. The above conversions allow usto integrate both solvers via a communication interface.
To reach an even tighter integration, we consider proof systems that combine resolution and polynomial calculus, i.e. the two most used proof systems in logic and algebraic solving. Based on such a proof system, which we call SRES, we introduce new types of solving algorithms that demostrate the synergy between Gröbner-like and DPLL-like solving. At the end of the second part of the dissertation, we provide some experiments based on a new benchmark which illustrate that the our new method based on DPLL has the potential to outperform CDCL SAT solvers.
In the third part of the thesis, we focus on practical attacks on various cryptograhic primitives. For instance, we apply SAT solvers in the case of algebraic fault attacks on the symmetric ciphers LED and derivatives of the block cipher AES. The main goal there is to derive so-called fault equations automatically from the hardware description of the cryptosystem and thus automatizate the attack. To give some extra power to a SAT solver that inverts the hash functions SHA-1 and SHA-2, we describe how to tweak the SAT solver using a programmatic interface such that the propagation of the solver and thus the attack itself is improved
Postcareer precarity : occupational challenges among former West African footballers in Northern Europe
No full textWhen Benoît Assou-Ekotto’s former coach Harry Redknapp announced that the Cameroonian international had decided to end his professional football career and become an adult movie actor, the story generated astonishment and amusement in social media networks and received widespread attention from the mass media. Although the French-born footballer denied the story’s validity shortly afterward, claiming Redknapp was joking, it entertained football fans and the wider public during the spring of 2017. However, while this was one of the very few occasions on which an African footballer’s postcareer trajectory was a matter of public debate, it also revealed that former African professional footballers’ whereabouts, occupations, and living conditions are widely unknown and remain an underresearched topic
Die kontemporäre universitäre Lehramtsausbildung für das Fach Sport in Deutschland und den USA - Eine explorative, melioristisch motivierte Vergleichsuntersuchung
No full textDie vergleichende Sportpädagogik als Teildisziplin und zugleich Schnittmenge der vergleichenden Erziehungswissenschaft und der allgemeinen Sportpädagogik macht es sich zur Aufgabe, sportbezogene Charakteristika zweier oder mehrerer Länder oder Kulturen miteinander zu vergleichen und den dadurch entstandenen Erkenntnisgewinn in verschiedensten Sphären nutzbar zu machen – eine Wissenschaftsdisziplin, wie auch Forschungsmethode, deren Möglichkeiten nicht nur für den Sportunterricht per se, sondern auch für die (universitäre) Ausbildung der zukünftigen Arrangeure des Sportunterrichts nutzbar gemacht werden können. Im Rahmen eines melioristisch motivierten Vergleichs macht es sich diese Forschungsarbeit zum Ziel, Unterschiede und Gemeinsamkeiten der universitären Sportlehramtsausbildung in Deutschland und den USA zu eruieren, deren Ursachen zu analysieren und auf dieser Basis Potenziale, Entwicklungsperspektiven und Handlungsoptionen für den positiven Fortschritt beider universitärer Ausbildungssysteme aufzuzeigen. Dabei werden sowohl außensystemische, wie auch innensystemische Untersuchungsaspekte berücksichtigt
Wirksamkeit des Flipped-Classroom-Konzepts in der Sekundarstufe
No full textResearch on flipped classroom instruction has substantially advanced in the past ten years. Flipped classroom refers to an instructional approach in which students study educational videos at home and do homework assignments in class. Since an increasing number of teachers wants to adopt the flipped classroom approach in their practice, further research—particularly in the context of secondary education—is clearly required. The two presented studies in this thesis aimed at examining the effectiveness of flipped classroom instruction in secondary education by conducting a meta-analytic synthesis of prior studies and an intervention study with a methodologically new approach. Specifically, the studies investigated whether and under which conditions the flipped classroom approach has a positive impact on student achievement and which learners benefit most from a flipped or video-based classroom.
In the first study, meta-analytic methods were used to examine whether the flipped classroom approach, after controlling for sampling error, positively effects student achievement in secondary education. Effect sizes were calculated for the research designs pre-test-post-test (Time), post-test only (PostOnly) and pre-test-post-test with control group (Treatment). Moreover, the impact of four moderator variables as boundary conditions of flipped classroom effectiveness was estimated: disciplinary field, length of the intervention, use of a quiz and use of a learning management system. The meta-analytical findings for the effect size Treatment confirmed the effectiveness of flipped classroom on student achievement in comparison to traditional instruction (Cohen’s d = 0.42). Moderator analyses on the effect size Time showed stronger effects for subjects in the STEM area (science, technology, engineering, mathematics) than for foreign languages and humanities. The effect sizes were also higher for shorter intervention studies than for longer ones and if quiz at home had been left out. Moderator analyses on the effect sizes PostOnly and Treatment made clear that the effect sizes for intervention studies without a learning management system were higher than with a learning management system.
The second study aimed to compare flipped classroom with other forms of video-based instruction and determine which types of students benefit most from video-based instruction. Thirty-eight school classes with 848 ninth-grade students took part in a quasi-experimental pre-post-test intervention study over the course of four weeks. Two independent variables were completely crossed resulting in four experimental conditions: video (at home vs. in class) and instructional method (student-centred vs. teacher-centred). Multilevel analyses revealed that all four experimental conditions were equally effective in promoting students’ learning gains. At-risk, average and excellent students profited least from video-based instruction. Confident and independent students had the highest learning gains from pre- to post-test. The study constitutes a first step towards a comprehensive evaluation of flipped classroom by using a better-controlled research design and may contribute to a more objective discussion about the positive effects of flipped classroom.Die Forschung zum Unterrichtskonzept „Flipped Classroom“ hat in den letzten zehn Jahren erhebliche Fortschritte gemacht. Flipped Classroom steht für einen Unterrichtsansatz, bei dem sich Schülerinnen und Schüler zuhause mithilfe von Erklärvideos neue Lerninhalte aneignen und im Unterricht anschließend dazu passende Aufgaben bearbeiten. Da immer mehr Lehrkräfte diesen Unterrichtsansatz in der Praxis anwenden möchten, sind weitere Forschungsarbeiten—insbesondere im Bereich der Sekundarstufe—notwendig. Die beiden in dieser Arbeit vorgestellten Studien zielten darauf ab, die Wirksamkeit des Unterrichtskonzepts „Flipped Classroom“ im Sekundarbereich zu untersuchen, indem eine meta-analytische Synthese früherer Studien und eine Interventionsstudie mit einem methodisch neuen Ansatz durchgeführt wurden. In den Studien wurde insbesondere untersucht, ob und unter welchen Bedingungen sich der Flipped-Classroom-Ansatz positiv auf die Lernleistungen von Schülerinnen und Schüler auswirkt und welche Lernenden am meisten vom videobasierten Unterricht profitieren.
In der ersten Studie wurden meta-analytische Methoden verwendet, um zu untersuchen, ob der Flipped-Classroom-Ansatz, nachdem eine Kontrolle des Stichprobenfehlers vorgenommen wurde, die Lernleistung von Schülerinnen und Schülern positiv beeinflusst. Die Effektstärken wurden für die Forschungsdesigns Vor-Test-Nach-Test („Zeit“), „nur Nach-Test“ und Vor-Test-Nach-Test mit Kontrollgruppe („Treatment“) berechnet. Darüber hinaus wurde der Einfluss von vier Moderatorvariablen als Randbedingungen für die Effektivität des Flipped-Classroom-Konzepts geschätzt: disziplinäres Feld, Dauer der Intervention, Verwendung eines Quiz und Verwendung eines Lernmanagementsystems. Die meta-analytischen Ergebnisse für die Effektstärke „Treatment“ bestätigten die Wirksamkeit des Flipped-Classroom-Konzepts auf die Lernleistung der Schülerinnen und Schüler im Vergleich zum traditionellen Unterricht (Cohen‘s d = 0.42). Moderatoranalysen der Effektstärken „Zeit“ zeigten stärkere Effekte für Fächer im MINT-Bereich (Mathematik, Informatik, Naturwissenschaften, Technik) als für fremdsprachliche und geisteswissenschaftliche Fächer. Die Effektstärken waren zudem bei kürzeren Interventionsstudien höher als bei längeren und wenn auf ein Lernquiz zuhause verzichtet wurde. Moderatoranalysen der Effektstärken „Nur Nachtest“ und „Treatment“ haben verdeutlicht, dass die Effektstärken höher ausfallen, wenn kein Lernmanagementsystem verwendet wird.
Die zweite Studie zielte darauf ab, das Flipped-Classroom-Konzept mit anderen Formen des videobasierten Unterrichts zu vergleichen und festzustellen, welche Schülertypen am meisten vom videobasierten Unterricht profitieren. 38 Schulklassen mit 848 Schülerinnen und Schülern der neunten Klasse nahmen über einen Zeitraum von vier Wochen an einer quasi-experimentellen Interventionsstudie mit Vor- und Nach-Test teil. Zwei unabhängige Variablen wurden vollständig gekreuzt, was zu vier experimentellen Bedingungen führte: Video (zuhause vs. im Unterricht) und Unterrichtsmethode (schülerzentriert vs. lehrerzentriert). Mehrebenenanalysen ergaben, dass alle vier experimentellen Bedingungen die Lerngewinne der Schülerinnen und Schüler gleichermaßen effektiv förderten. Risikoschülerinnen und Risikoschüler sowie durchschnittliche und ausgezeichnete Schülerinnen und Schüler profitierten am wenigsten vom videobasierten Unterricht. Selbstbewusste und selbstständige Schülerinnen und Schüler hatten die höchsten Lerngewinne von Vor- zu Nach-Test. Die Studie stellt einen ersten Schritt in Richtung einer umfassenderen Bewertung des Flipped-Classroom-Konzepts unter Verwendung eines besser kontrollierten Forschungsdesigns dar und könnte zu einer objektiveren Diskussion über die positiven Auswirkungen des Flipped-Classroom-Konzepts beitragen