Filtern
Erscheinungsjahr
- 2020 (30) (entfernen)
Dokumenttyp
- Dissertation (17)
- Masterarbeit (6)
- Ausgabe (Heft) zu einer Zeitschrift (3)
- Bachelorarbeit (2)
- Wissenschaftlicher Artikel (1)
- Preprint (1)
Sprache
- Englisch (30) (entfernen)
Schlagworte
- model-based (2)
- Abwasserreinigung (1)
- Aquatic Ecotoxicology (1)
- Artificial Intelligence (1)
- BPMN (1)
- Beschichtung (1)
- Biohydrogel (1)
- Biopolymere (1)
- Bombina variegata (1)
- Carry-over effects (1)
- Climate (1)
- Cognitive functions (1)
- Conservation (1)
- DMN (1)
- Data manipulation (1)
- Data protection (1)
- Datenschutz (1)
- Demographie (1)
- Demography (1)
- Ecotoxicology (1)
- Elastic net (1)
- Environmental factors (1)
- Erzieher (1)
- Erzieherin (1)
- Evidence-based Psychotherapy (1)
- Fast-slow continuum (1)
- GDPR (1)
- Gel effect (1)
- Gelbbauchunke (1)
- Genetics (1)
- Genetik (1)
- Habitat Fragmentation (1)
- Habitatfragmentierung (1)
- IPT (1)
- Klima (1)
- Kohlenstoffschichten (1)
- Langlebigkeit (1)
- Life history (1)
- Longevity (1)
- Machine-Learning (1)
- Machinelles lernen (1)
- Material Point Method (1)
- Mixed method (1)
- Mucilage (1)
- NMR relaxometry (1)
- Nanoparticles (1)
- Nanopartikel (1)
- Native language identification (1)
- Naturschutzmanagement (1)
- Oberflächenveredelung (1)
- Personality (1)
- Persönlichkeit (1)
- Pesticides (1)
- Pestizid (1)
- Physik (1)
- Physiksimulation (1)
- Pollinators (1)
- Process Quality (1)
- Prozessqualität (1)
- Pädagogik (1)
- Railway Research (1)
- Railway Research Topics (1)
- Railway Safety (1)
- Railway Safety Research (1)
- Random Forest (1)
- Recovery (1)
- Rehabilitation (1)
- Rheinland-Pfalz (1)
- Rhineland-Palatinate (1)
- Rhizosphere (1)
- Rückverfolgbarkeit (1)
- Salinisation (1)
- Sand (1)
- Schizophrenie (1)
- Schnee (1)
- Sozialpsychologie (1)
- Species turnover (1)
- UML (1)
- Ubuntu (1)
- Verification (1)
- Versalzung (1)
- Wastewater treatment plants (1)
- Wirbellose (1)
- Yellow-bellied toad (1)
- amorphous hydrogenated carbon layer (1)
- blockchain (1)
- carbon hybridisation (1)
- competence- and control beliefs (1)
- conflict detection (1)
- core self-evaluations (1)
- data protection (1)
- data sharing (1)
- delivery drone (1)
- density separation (1)
- digestion (1)
- distributed ledger (1)
- drone (1)
- emic-etic (1)
- fairness (1)
- healthcare (1)
- mixtures (1)
- monitoring (1)
- pesticides (1)
- plastic debris (1)
- privacy by design (1)
- privacy impact assessment (1)
- risks (1)
- sample pretreatment (1)
- self-efficacy (1)
- soil organic matter (1)
- streams (1)
- technology acceptance model (1)
- traceability (1)
- transformation (1)
Institut
- Fachbereich 7 (5)
- Institut für Computervisualistik (5)
- Institut für Management (4)
- Fachbereich 8 (3)
- Institut für Integrierte Naturwissenschaften, Abt. Biologie (3)
- Fachbereich 4 (2)
- Institut für Softwaretechnik (2)
- Institut für Umweltwissenschaften (2)
- Institute for Web Science and Technologies (2)
- Institut für Informatik (1)
- Institut für Integrierte Naturwissenschaften, Abt. Physik (1)
- Institut für Psychologie (1)
- Institut für Wirtschafts- und Verwaltungsinformatik (1)
Social-Media Plattformen wie Twitter oder Reddit bieten Nutzern nahezu ohne Beschränkungen die Möglichkeit, ihre Meinungen über aktuelle Ereignisse zu veröffentlichen, diese mit anderen zu teilen und darüber zu diskutieren. Während die Mehrheit der Nutzer diese Plattformen nur als reines Diskussionsportal verwenden, gibt es jedoch Nutzergruppen, welche aktiv und gezielt versuchen, diese veröffentlichten Meinungen in ihrem Sinne zu beeinflussen bzw. zu manipulieren. Durch wiederholtes Verbreiten von bearbeiteten Fake-News oder stark polarisierenden Meinungen im gesamten politischen Spektrum können andere Nutzer beeinflusst, manipuliert und unter Umständen zum Träger von Hassreden und extremen politischen Positionen werden. Viele dieser Nutzergruppen sind vor allem in englischsprachigen Portalen anzutreffen, in denen sie sich überwiegend als Muttersprachler ausgeben. In dieser Arbeit stellen wir eine Methode vor, englische Muttersprachler und Nicht-Muttersprachler, die Englisch als Fremdsprache verwenden, anhand von ausgewählten englischen Social Media Texten zu unterscheiden. Dazu implementieren wir textmerkmalbasierte Modelle, welche für traditionelle Machine-Learning Prozesse und neuartigen AutoML-Pipelines zur Klassifizierung von Texten verwendet werden. Wir klassifizieren dabei Sprachfamilie, Muttersprache und Ursprung eines beliebigen englischen Textes. Die Modelle werden an einem bestehenden Datensatz von Reddit, welcher hauptsächlich aus englischen Texten von europäischen Nutzern besteht, und einem neu erstellten Twitter Datensatz, der Tweets von aktuellen Themen in verschiedenen Ländern enthält, angewandt. Wir evaluieren dabei vergleichsweise die erhaltenen Resultate unserer Pipeline zu traditionellen Maschinenlernprozessen zur Texterkennung anhand von Präzision, Genauigkeit und F1-Maßen der Vorhersagen. Wir vergleichen zudem die Ergebnisse auf Unterschiede der Sprachnutzung auf den unterschiedlichen Plattformen sowie den ausgewählten Themenbereichen. Dabei erzielen wir eine hohe Vorhersagewahrscheinlichkeit für alle gewählten Kategorien des erstellten Twitter Datensatzes und stellen unter anderem eine hohe Abweichung in Bezug auf die durchschnittliche Textlänge insbesondere bei Nutzern aus dem baltoslawischen Sprachraum fest.
Since the invention of U-net architecture in 2015, convolutional networks based on its encoder-decoder approach significantly improved results in image analysis challenges. It has been proven that such architectures can also be successfully applied in different domains by winning numerous championships in recent years. Also, the transfer learning technique created an opportunity to push state-of-the-art benchmarks to a higher level. Using this approach is beneficial for the medical domain, as collecting datasets is generally a difficult and expensive process.
In this thesis, we address the task of semantic segmentation with Deep Learning and make three main contributions and release experimental results that have practical value for medical imaging.
First, we evaluate the performance of four neural network architectures on the dataset of the cervical spine MRI scans. Second, we use transfer learning from models trained on the Imagenet dataset and compare it to randomly initialized networks. Third, we evaluate models trained on the bias field corrected and raw MRI data. All code to reproduce results is publicly available online.
The European landscape is dominated by intensive agriculture which leads to widespread impact on the environment. The frequent use of agricultural pesticides is one of the major causes of an ongoing decline in flower-visiting insects (FVIs). The conservation of this ecologically diverse assemblage of mobile, flying insect species is required by international and European policy. To counteract the decrease in species numbers and their abundances, FVIs need to be protected from anthropogenic stressors. European pesticide risk assessment was devised to prevent unacceptable adverse consequences of pesticide use on FVIs. However, there is an ongoing discussion by scientists and policy-makers if the current risk assessment actually provides adequate protection for FVI species.
The first main objective of this thesis was to investigate pesticide impact on FVI species. The scientific literature was reviewed to identify groups of FVIs, summarize their ecology, and determine their habitat. This was followed by a synthesis of studies about the exposure of FVIs in their habitat and subsequent effects. In addition, the acute sensitivity of one FVI group, bee species, to pesticides was studied in laboratory experiments.
The second main objective was to evaluate the European risk assessment for possible deficits and propose improvements to the current framework. Regulatory documents were screened to assess the adequacy of the guidance in place in light of the scientific evidence. The suitability of the honey bee Apis mellifera as the currently only regulatory surrogate species for FVIs was discussed in detail.
The available scientific data show that there are far more groups of FVIs than the usually mentioned bees and butterflies. FVIs include many groups of ecologically different species that live in the entire agricultural landscape. Their habitats in crops and adjacent semi-natural areas can be contaminated by pesticides through multiple pathways. Environmentally realistic exposure of these habitats can lead to severe effects on FVI population parameters. The laboratory studies of acute sensitivity in bee species showed that pesticide effects on FVIs can vary greatly between species and pesticides.
The follow-up critical evaluation of the European FVI risk assessment revealed major shortcomings in exposure and effect assessment. The honey bee proved to be a sufficient surrogate for bee species in lower tier risk assessment. Additional test species may be chosen for higher tier risk assessment to account for ecological differences. This thesis shows that the ecology of FVIs should generally be considered to a greater extent to improve the regulatory process. Data-driven computational approaches could be used as alternative methods to incorporate ecological trait data in spatio-temporal scenarios. Many open questions need to be answered by further research to better understand FVI species and promote necessary changes to risk assessment. In general, other FVI groups than bees need to be investigated. Furthermore, comprehensive data on FVI groups and their ecology need to be collected. Contamination of FVI habitat needs to be linked to exposure of FVI individuals and ecologically complex effects on FVI populations should receive increased attention. In the long term, European FVI risk assessment would benefit from shifting its general principles towards more scientifically informed regulatory decisions. This would require a paradigm shift from arbitrary assumptions and unnecessarily complicated schemes to a substantiated holistic framework.
Initial goal of the current dissertation was the determination of image-based biomarkers sensitive for neurodegenerative processes in the human brain. One such process is the demyelination of neural cells characteristic for Multiple sclerosis (MS) - the most common neurological disease in young adults for which there is no cure yet. Conventional MRI techniques are very effective in localizing areas of brain tissue damage and are thus a reliable tool for the initial MS diagnosis. However, a mismatch between the clinical fndings and the visualized areas of damage is observed, which renders the use of the standard MRI diffcult for the objective disease monitoring and therapy evaluation. To address this problem, a novel algorithm for the fast mapping of myelin water content using standard multiecho gradient echo acquisitions of the human brain is developed in the current work. The method extents a previously published approach for the simultaneous measurement of brain T1, T∗ 2 and total water content. Employing the multiexponential T∗ 2 decay signal of myelinated tissue, myelin water content is measured based on the quantifcation of two water pools (myelin water and rest) with different relaxation times. Whole brain in vivo myelin water content maps are acquired in 10 healthy controls and one subject with MS. The in vivo results obtained are consistent with previous reports. The acquired quantitative data have a high potential in the context of MS. However, the parameters estimated in a multiparametric acquisition are correlated and constitute therefore an ill-posed, nontrivial data analysis problem. Motivated by this specific problem, a new data clustering approach is developed called Nuclear Potential Clustering, NPC. It is suitable for the explorative analysis of arbitrary dimensional and possibly correlated data without a priori assumptions about its structure. The developed algorithm is based on a concept adapted from nuclear physics. To partition the data, the dynamic behavior of electrically even charged nucleons interacting in a d-dimensional feature space is modeled. An adaptive nuclear potential, comprised of a short-range attractive (Strong interaction) and a long-range repulsive term (Coulomb potential), is assigned to each data point. Thus, nucleons that are densely distributed in space fuse to build nuclei (clusters), whereas single point clusters are repelled (noise). The algorithm is optimized and tested in an extensive study with a series of synthetic datasets as well as the Iris data. The results show that it can robustly identify clusters even when complex configurations and noise are present. Finally, to address the initial goal, quantitative MRI data of 42 patients are analyzed employing NPC. A series of experiments with different sets of image-based features show a consistent grouping tendency: younger patients with low disease grade are recognized as cohesive clusters, while those of higher age and impairment are recognized as outliers. This allows for the definition of a reference region in a feature space associated with phenotypic data. Tracking of the individual's positions therein can disclose patients at risk and be employed for therapy evaluation.
Although most plastic pollution originates on land, current research largely remains focused on aquatic ecosystems. Studies pioneering terrestrial microplastic research have adapted analytical methods from aquatic research without acknowledging the complex nature of soil. Meanwhile, novel methods have been developed and further refined. However, methodical inconsistencies still challenge a comprehensive understanding of microplastic occurrence and fate in and on soil. This review aims to disentangle the variety of state-of-the-art sample preparation techniques for heterogeneous solid matrices to identify and discuss best-practice methods for soil-focused microplastic analyses. We show that soil sampling, homogenization, and aggregate dispersion are often neglected or incompletely documented. Microplastic preconcentration is typically performed by separating inorganic soil constituents with high-density salt solutions. Not yet standardized but currently most used separation setups involve overflowing beakers to retrieve supernatant plastics, although closed-design separation funnels probably reduce the risk of contamination. Fenton reagent may be particularly useful to digest soil organic matter if suspected to interfere with subsequent microplastic quantification. A promising new approach is extraction of target polymers with organic solvents. However, insufficiently characterized soils still impede an informed decision on optimal sample preparation. Further research and method development thus requires thorough validation and quality control with well-characterized matrices to enable robust routine analyses for terrestrial microplastics.
Bäche gehören zu den gefährdetsten Ökosystemen, da sie diversen anthropogenen Stressoren ausgesetzt sind, wobei Pestizide für landwirtschaftliche Bäche am relevantesten erscheinen. Aufgrund der Diskrepanz zwischen modellierten und gemessenen Pestizid-konzentrationen ist Monitoring nötig um zukünftige Risikobewertungen und Zulassungen zu verbessern. Festzustellen ob biotische Stressreaktionen über große räumliche und zeitliche Skalen ähnlich sind, ist nötig um Schwellenwerte zum Schutz vor Stressoren abzuschätzen.
Diese Doktorarbeit beginnt mit einem Überblick über Pestizideintrittspfade in Bäche, sowie dem momentanen Stand des Pestizidmonitorings gefolgt von der Zielsetzung der Doktorarbeit. In Kapitel 2 werden Ergebnisse aus Schöpfproben von Routinemonitoring mehrerer Länder analysiert um die häufigsten Pestizidmischungen zu identifizieren. Diese Mischungen werden von wenigen Pestiziden gebildet, wobei Herbizide dominieren. Die nachgewiesenen Mischungen unterscheiden sich regional, da Nachweisgrenzen und Stoffumfang variieren. Aktuelles Routinemonitoring umfasst bisher keine Probenahmen während durch Starkregenereignisse hervorgerufene Pestizidspitzen, die wahrscheinlich Pestizidmischungen beeinflussen. In Kapitel 3 werden Sammelraten für 42 Pestizide bei der Benutzung von Passiv-sammlern vorgestellt und Empfehlungen zum Monitoring von feldrelevanten Pestizidspitzen gegeben. Damit konnte in Kapitel 4 ein Pestizidgradient in einer osteuropäischen Region aufgestellt werden in der die Landwirtschaftsintensität von niedrig bis hoch reicht. Dabei wurden Regenereignisse beprobt und eine Vielzahl von Pestiziden analysiert. Dies führte zu vielen gleichzeitig nachgewiesenen Pestiziden, von denen ein bis drei die Pestizidtoxizität bestimmten. Diese zeigte jedoch keinen Zusammenhang zur landwirtschaftlichen Intensität. Durch Mikrokosmenexperimente wurde in Kapitel 5 die Stressantwort von Pilzgemeinschaften, den Hyphomyceten, und deren assoziierter Ökosystemfunktion des Laubabbaus untersucht. Effekte einer feldrelevanten Fungizidmischung wurde über drei biogeographische Regionen sowie drei aufeinanderfolgende Zyklen von mikrobieller Laubkolonisation und -abbau untersucht. Trotz anfänglich unterschiedlichen Gemeinschaften waren Stressantworten sowie Erholungen in den untersuchten Regionen ähnlich, was auf ein generelles Muster hindeutet.
Insgesamt trägt diese Doktorarbeit zum verbesserten Verständnis von Vorkommen und Konzentrationen von Pestizidmischungen, deren Monitoring sowie ihren Auswirkungen auf eine Ökosystemfunktion bei. Wir konnten zeigen, dass die abgeschätzten Pestizidtoxizitäten potentiell Nichtzielorganismen und somit ganze Ökosystem beeinflussen. Routinemonitoring unterschätzt diese Gefahr bisher jedoch wahrscheinlich. Effekte, welche Verluste in Biodiversität sowie Funktionen hervorrufen, können verringert werden indem zugelassene Pestizide mit anhaltendem Monitoring neu bewertet werden und die Datenlage zu Pestizidwirkungen verbessert wird.
Railway safety is a topic which gains the public attention only if major railway accidents happen. This is because railway is considered as a safe mode of travel by the public. However, to ensure the safety of the railway system railway companies as well as universities conduct a broad spectrum of research. An overview of this research has not yet been provided in the scholarly literature. Therefore, this thesis follows two objectives. First an overview and ranking of railway safety research universities should be provided. Second, based on these universities, it should be identified which are the most relevant and influential research topics. The ranking is based on the research method “literature review” which forms the methodical basis for this thesis. To evaluate the universities based on a measurable and objective criterion, the number of citations of the researchers from each university is gathered. As a result, the University of Leuven for the civil engineering, Milan Politechnico for mechanical enginering and the University of Loughborough for electrical engineering are identified as the leading university in their field of railway safety research. The top universities for each discipline are distributed all over Europe, North America and Asia. However, a clear focus on the US and British universities is observed. For identification of the most relevant and influential topics the keywords from the publications which are considered in the ranking procedure are analyzed. Focus areas among these keywords are revealed by calculating the count of each keyword. High-speed trains as well as maintenance are recognized as the highly relevant topics in both civil and mechanical engineering. Furthermore, the topic of railway dynamics for mechanical engineering and noise and vibration for civil engineering are identified as the leading topics in the respective discipline. Achieving both research goals required exploratory approaches. Therefore, this thesis leaves open space for future research to deepen the individual topics which are approached in each section. A validation of the results through experts interviews as well as a deepening of the analysis through increasing the number of analyzed universities as well as applying statistical methods is recommended.
The three biodegradable polymers polylactic acid (PLA), polyhydroxybutyrate (PHB) and polybutylene adipate terephthalate (PBAT) were coated with hydrogenated amorphous carbon layers (a-C:H) in the context of this thesis. A direct alignment of the sample surface to the source was chosen, resulting in the deposition of a robust, r-type a-C:H. At the same time, a partly covered silicon wafer was placed together with the polymers in the coating chamber and was coated. Silicon is a hard material and serves as a reference for the applied layers. Due to the hardness of the material, no mixed phase occurs between the substrate and the applied layer (no interlayer formation). In addition, the thickness of the applied layer can be estimated with the help of the silicon sample.
The deposition of the layer was realized by radio frequency plasma enhanced chemical vapor deposition (RF-PECVD). For the coating the samples were pre-treated with an oxygen plasma. Acetylene was used as precursor gas for the plasma coating. Coatings with increasing thickness in 50 nm steps from 0-500 nm were realised.
The surface analysis was performed using several techniques: The morphology and layer stability were analyzed with scanning electron microscopy (SEM) measurements. The wettability was determined by contact angle technique. In addition, the contact angles provide macroscopic information about the bond types of the carbon atoms present on the surface. For microscopic analysis of the chemical composition of the sample and layer surfaces, diffuse reflectance Fourier transform infrared spectroscopy (DRIFT) as well as synchrotron based X-ray photon spectroscopy (XPS) and near edge X-ray absorption fine structure spectroscopy (NEXAFS) were used.
All coated polymers showed several cases of layer failure due to internal stress in the layers. However, these were at different layer thicknesses, so there was a substrate effect. In addition, it is visible in the SEM images that the coatings of PLA and PHB can cause the applied layer to wave, the so-called cord buckling. This does not occur with polymer PBAT, which indicates a possible better bonding of the layer to the polymer. The chemical analyses of the layer surfaces show for each material a layer thickness dependent ratio of sp² to sp³ bonds of carbon, which alternately dominate the layer. In all polymers, the sp³ bond initially dominates, but the sp² to sp³ ratio changes at different intervals. Although the polymers were coated in the same plasma, i.e. the respective layer thicknesses (50 nm, 100 nm, ...) were applied in the same plasma process, the respective systems differed considerably from each other. A substrate effect is therefore demonstrably present. In addition, it was found that a change in the dominant bond from sp³ to sp² is an indication ofan upcoming layer failure of the a-C:H layer deposited on the polymer. In the case of PLA, this occurs immediately with change to sp² as the dominant bond; in the case of PHB and PBAT, this occurs with different delay to increased layer thicknesses (at PHB 100 nm, at PBAT approx. 200 nm.
Overall, this thesis shows that there is a substrate effect in the coating of the biodegradable polymers PLA, PHB and PBAT, since despite the same coating there is a different chemical composition of the surface at the respective layer thicknesses. In addition, a layer failure can be predicted by analyzing the existing bond.
Data-minimization and fairness are fundamental data protection requirements to avoid privacy threats and discrimination. Violations of data protection requirements often result from: First, conflicts between security, data-minimization and fairness requirements. Second, data protection requirements for the organizational and technical aspects of a system that are currently dealt with separately, giving rise to misconceptions and errors. Third, hidden data correlations that might lead to influence biases against protected characteristics of individuals such as ethnicity in decision-making software. For the effective assurance of data protection needs,
it is important to avoid sources of violations right from the design modeling phase. However, a model-based approach that addresses the issues above is missing.
To handle the issues above, this thesis introduces a model-based methodology called MoPrivFair (Model-based Privacy & Fairness). MoPrivFair comprises three sub-frameworks: First, a framework that extends the SecBPMN2 approach to allow detecting conflicts between security, data-minimization and fairness requirements. Second, a framework for enforcing an integrated data-protection management throughout the development process based on a business processes model (i.e., SecBPMN2 model) and a software architecture model (i.e., UMLsec model) annotated with data protection requirements while establishing traceability. Third, the UML extension UMLfair to support individual fairness analysis and reporting discriminatory behaviors. Each of the proposed frameworks is supported by automated tool support.
We validated the applicability and usability of our conflict detection technique based on a health care management case study, and an experimental user study, respectively. Based on an air traffic management case study, we reported on the applicability of our technique for enforcing an integrated data-protection management. We validated the applicability of our individual fairness analysis technique using three case studies featuring a school management system, a delivery management system and a loan management system. The results show a promising outlook on the applicability of our proposed frameworks in real-world settings.
Schizophrenia is a chronic mental health disorder, which changes rapidly the life of the persons and their families, who suffer from it. It causes high biological and psychological vulnerability as well as cognitive, emotional and behavioral disorders. Nowadays, evidence-based pharmacotherapy and psychotherapy are available aiming the rehabilitation and recovery of individuals with schizophrenia. A democratic society is obliged to give these people the opportunity to have an access to those treatments.
The following three published studies present this dissertation thesis and have a common focus on the implementation of evidence-based psychotherapy in individuals with schizophrenia.
The first study evaluates the efficacy of the Integrated Psychological Therapy (IPT) in Greece, one of the most evaluated rehabilitation programs. IPT was compared to
Treatment as Usual (TAU) in a randomized controlled trial (RCT) with 48 individuals with schizophrenia. Significant effects favouring IPT were found in working memory,
in social perception, in negative symptoms, in general psychopathology and in insight. This study supports evidence for the efficacy of IPT in Greece.
The second study evaluates a second hypothesis, when IPT is more and less effective regarding treatment resistant schizophrenia (TRS) and non treatment resistant
schizophrenia (NTRS). It is a part of the first paper. Significant effects favouring NTRS were found for verbal memory, for symptoms, for functioning and quality of
life. Effect sizes showed superiority of NTRS in comparison to TRS. IPTTRS showed on the other side some significant improvements. This study presents the initial findings of a larger study to be conducted internationally for the first time.
The third study is a systematic review, which aims to evaluate the efficacy of Cognitive Behavioral Therapy (CBT), of Meta Cognitive Therapy (MCT), Metacognitive Training (MCTR), Metacognitive Reflection and Insight Therapy
(MERIT), of various Rehabilitation Programs and Recovery Programs in individuals with schizophrenia. 41 RCTs and 12 Case Studies were included. The above interventions are efficacious in the improvement of cognitions, symptoms, functional outcome, insight, self-esteem, comorbid disorders and metacognitive capacity.
The three studies provide insight regarding the importance of evidence-based psychotherapy in persons with schizophrenia leading to recovery and reintegration into
society. Future RCTs with larger samples and long-term follow up, combining evidence-based psychotherapies for individuals with schizophrenia need to be done.
Konstituenten-Parsing versucht, syntaktische Struktur aus einem Satz zu extrahieren. Diese Parsing-Systeme sind in vielen maschinellen Sprachverarbeitungsanwendungen hilfreich, wie z.B. bei der Grammatikprüfung, der Beantwortung von Fragen und der Informationsextraktion. In dieser Masterarbeit geht es um die Implementierung eines Konstituentenparsers für die deutsche Sprache mit Hilfe von neuronalen Netzen. In der Vergangenheit wurden wiederkehrende neuronale Netze beim Aufbau eines Parsers und auch bei vielen maschinellen Sprachverarbeitungsanwendungen verwendet. Dabei werden Module des neuronalen Netzes mit Selbstaufmerksamkeit intensivgenutzt, um Sätze effektiv zu verstehen. Bei mehrschichtigen Selbstaufmerksamkeitsnetzwerken erreicht das konstituierende
Parsen 93,68% F1-Scoret. Dies wird noch weiter verbessert, indem sowohl Zeichen- als auch Worteinbettungen als Darstellung des Inputs verwendet werden. Ein F1-Score von 94,10% wurde am besten durch den Konstituenten-Parser erreicht, der nur den bereitgestellten Datensatz verwendet. Mit Hilfe externer Datensätze wie der deutschen Wikipedia werden vortrainierte ELMo-Modelle zusammen mit Selbstbeobachtungsnetzwerken verwendet, die einen F1-Score von 95,87% erreichen.
Die Material Point Method (MPM) hat sich in der Computergrafik als äußerst fähige Simulationsmethode erwiesen, die in der Lage ist ansonsten schwierig zu animierende Materialien zu modellieren [1, 2]. Abgesehen von der Simulation einzelner Materialien stellt die Simulation mehrerer Materialien und ihrer Interaktion weitere Herausforderungen bereit. Dies ist Thema dieser Arbeit. Es wird gezeigt, dass die MPM durch die Fähigkeit Eigenkollisionen implizit handzuhaben ebenfalls in der Lage ist Kollisionen zwischen Objekten verschiedenster Materialien zu beschreiben, selbst, wenn verschiedene Materialmodelle eingesetzt werden. Dies wird dann um die Interaktion poröser Materialien wie in [3] erweitert, was ebenfalls gut mit der MPM integriert. Außerdem wird gezeigt das MPM auf Basis eines einzelnen Gitters als Untermenge dieses Mehrgitterverfahrens betrachtet werden kann, sodass man das gleiche Verhalten auch mit mehreren Gittern modellieren kann. Die poröse Interaktion wird auf beliebige Materialien erweitert, einschließlich eines frei formulierbaren Materialinteraktionsterms. Das Resultat ist ein flexibles, benutzersteuerbares Framework das unabhängig vom Materialmodell ist. Zusätzlich wird eine einfache GPU-Implementation der MPM vorgestellt, die die Rasterisierungspipeline benutzt um Schreibkonflikte aufzulösen. Anders als andere Implementationen wie [4] ist die vorgestellte Implementation kompatibel mit einer Breite an Hardware.
Environmental processes transforming inorganic nanoparticles: implications on aquatic invertebrates
(2020)
Engineered inorganic nanoparticles (EINPs) are produced and utilized on a large scale and will end up in surface waters. Once in surface waters, EINPs are subjected to transformations induced by environmental processes altering the particles’ fate and inherent toxicity. UV irradiation of photoactive EINPs is defined as one effect-inducing pathway, leading to the formation of reactive oxygen species (ROS), increasing EINP toxicity by exerting oxidative stress in aquatic life. Simultaneously, UV irradiation of photoactive EINP alters the toxicity of co-occurring micropollutants (e.g. pesticides) by affecting their degradation. The presence of natural organic matter (NOM) reduces the agglomeration and sedimentation of EINPs, extending the exposure of pelagic species, while delaying the exposure of benthic species living in and on the sediment, which is suggested as final sink for EINPs. However, the joint impact of NOM and UV irradiation on EINP-induced toxicity, but also EINP-induced degradation of micropollutants, and the resulting risk for aquatic biota, is poorly understood. Although potential effects of EINPs on benthic species are increasingly investigated, the importance of exposure pathways (waterborne or dietary) is unclear, along with the reciprocal pathway of EINPs, i.e. the transport back from aquatic to terrestrial ecosystems. Therefore, this thesis investigates: (i) how the presence of NOM affects the UV-induced toxicity of the model EINP titanium dioxide (nTiO2) on the pelagic organism Daphnia magna, (ii) to which extent UV irradiation of nTiO2 in the presence and absence of NOM modifies the toxicity of six selected pesticides in D. magna, (iii) potential exposure pathway dependent effects of nTiO2 and silver (nAg) EINPs on the benthic organism Gammarus fossarum, and (iv) the transport of nTiO2 and gold EINPs (nAu) via the merolimnic aquatic insect Chaetopteryx villosa back to terrestrial ecosystems. nTiO2 toxicity in D. magna increased up to 280-fold in the presence of UV light, and was mitigated by NOM up to 12-fold. Depending on the pesticide, UV irradiation of nTiO2 reduced but also enhanced pesticide toxicity, by (i) more efficient pesticide degradation, and presumably (ii) formation of toxic by-products, respectively. Likewise, NOM reduced and increased pesticide toxicity, induced by (i) protection of D. magna against locally acting ROS, and (ii) mitigation of pesticide degradation, respectively. Gammarus’ energy assimilation was significantly affected by both EINPs, however, with distinct variation in direction and pathway dependence between nTiO2 and nAg. EINP presence delayed C. villosa emergence by up to 30 days, and revealed up to 40% reduced lipid reserves, while the organisms carried substantial amounts of nAu (~1.5 ng/mg), and nTiO2 (up to 2.7 ng/mg). This thesis shows, that moving test conditions of EINPs towards a more field-relevant approach, meaningfully modifies the risk of EINPs for aquatic organisms. Thereby, more efforts need to be made to understand the relative importance of EINP exposure pathways, especially since a transferability between different types of EINPs may not be given. When considering typically applied risk assessment factors, adverse effects on aquatic systems might already be expected at currently predicted environmental EINP concentrations in the low ng-µg/L range.
Bio-medical data comes in various shapes and with different representations.
Domain experts use such data for analysis or diagnosis,
during research or clinical applications. As the opportunities to obtain
or to simulate bio-medical data become more complex and productive,
the experts face the problem of data overflow. Providing a
reduced, uncluttered representation of data, that maintains the data’s
features of interest falls into the area of Data Abstraction. Via abstraction,
undesired features are filtered out to give space - concerning the
cognitive and visual load of the viewer - to more interesting features,
which are therefore accentuated. To address this challenge, the dissertation
at hand will investigate methods that deal with Data Abstraction
in the fields of liver vasculature, molecular and cardiac visualization.
Advanced visualization techniques will be applied for this purpose.
This usually requires some pre-processing of the data, which will also
be covered by this work. Data Abstraction itself can be implemented
in various ways. The morphology of a surface may be maintained,
while abstracting its visual cues. Alternatively, the morphology may
be changed to a more comprehensive and tangible representation.
Further, spatial or temporal dimensions of a complex data set may
be projected to a lower space in order to facilitate processing of the
data. This thesis will tackle these challenges and therefore provide an
overview of Data Abstraction in the bio-medical field, and associated
challenges, opportunities and solutions.
This thesis examined two specific cases of point and diffuse pollution, pesticides and salinisation, which are two of the most concerning stressors of Germany’s freshwater bodies. The findings of this thesis were organized into three major components, of which the first component presents the contribution of WWTPs to pesticide toxicity (Chapter 2). The second component focuses on the current and future background salt ion concentrations under climate change with the absence of anthropogenic activities (Chapter 3). Finally, the third major component shows the response of invertebrate communities in terms of species turnover to levels of salinity change, considered as a proxy for human-driven salinisation (Chapter 4).
Molecular dynamics (MD) as a field of molecular modelling has great potential to revolutionize our knowledge and understanding of complex macromolecular structures. Its field of application is huge, reaching from computational chemistry and biology over material sciences to computer-aided drug design. This thesis on one hand provides insights into the underlying physical concepts of molecular dynamics simulations and how they are applied in the MD algorithm, and also briefly illustrates different approaches, as for instance the molecular mechanics and molecular quantum mechanics approaches.
On the other hand an own all-atom MD algorithm is implemented utilizing and simplifying a version of the molecular mechanics based AMBER force field published by \big[\cite{cornell1995second}\big]. This simulation algorithm is then used to show by the example of oxytocin how individual energy terms of a force field function. As a result it has been observed, that applying the bond stretch forces alone caused the molecule to be compacted first in certain regions and then as a whole, and that with adding more energy terms the molecule got to move with increasing flexibility.
Despite widespread plans of big companies like Amazon and Google to develop unmanned delivery drones, scholarly research in this field is scarce, especially in the information systems field. From technical and legal perspectives, drone delivery in last-mile scenarios is in a quite mature state. However, estimates of user acceptance are varying between high skepticism and exaggerated optimism. This research follows a mixed method approach consisting both qualitative and quantitative research, to identify and test determinants of consumer delivery drone service adoption. The qualitative part rests on ten interviews among average consumers, who use delivery services on a regular basis. Insights gained from the qualitative part were used to develop an online survey and to assess the influence of associated risks on adoption intentions. The quantitative results show that especially financial and physical risks impede drone delivery service adoption. Delivery companies who are currently thinking about providing a delivery drone service may find these results useful when evaluating usage behaviors in the future market for delivery drones.
The history of human kind is characterized by social conflict. Every conflict can be the starting point of social change or the escalation into more destructive forms. The social conflict in regard to rising numbers of refugees and their acceptance that arose in most host countries in 2015 already took on destructive forms – in Germany, right-wing extremists attacked refugee shelters and even killed multiple people, including political leaders who openly supported refugees. Thus, incompatible expectancies and values of different parts of the society led to violent action tendencies, which tremendously threaten intergroup relations. Psychological research has developed several interventions in past decades to improve intergroup relations, but they fall short, for example, when it comes to the inclusion of people with extreme attitudes and to precisely differentiate potential prosocial outcomes of the interventions. Thus, this dissertation aimed to a) develop psychological interventions, that could also be applied to people with more extreme attitudes, thereby putting a special emphasis on collecting a diverse sample; b) gain knowledge about target- and outcome specific effects: Who benefits from which intervention and how can specific prosocial actions be predicted in order to develop interventions that guide needs-based actions; and c) shed light on potential underlying mechanisms of the interventions.
The dissertation will be introduced by the socio-political background that motivated the line of research pursued, before providing an overview of the conceptualization of social conflicts and potential psychological inhibitors and catalyzers for conflict transformation. Based on past research on socio-psychological interventions and their limitations, the aims of the dissertation will be presented in more detail, followed by a short summary of each manuscript. Overall, the present thesis comprises four manuscripts that were summarized in the general discussion into a road map for social-psychological interventions to put them into a broader perspective. The road map aspires to provide recommendations for increasing – either approach-oriented or support-oriented actions – by the socio-psychological interventions for a variety of host society groups depending on their pre-existing attitude towards refugees.
A Paradoxical Intervention targeting central beliefs of people with negative attitudes towards refugees influenced inhibitory and catalyzing factors for conflict transformation over the course of three experiments – thereby providing an effective tool to establish approach-oriented action tendencies, such as the willingness to get in contact with refugees. Further, the dissertation presents a novel mechanism – namely Cognitive Flexibility – which could explain the Paradoxical Interventions’ effect of past research. By positively affecting a context-free mindset, the Paradoxical Intervention could impact more flexible thought processes in general, irrespective of the topic tackled in the Paradoxical Intervention itself. For people with rather positive attitudes addressing emotions may increase specific support-oriented action tendencies. The dissertation provides evidence of a positive relation between moral outrage and hierarchy-challenging actions, such as solidarity-based collective action, and sympathy with prosocial hierarchy-maintaining support-oriented actions, such as dependency-oriented helping. These exclusive relations between specific emotions and action intentions provide important implications for the theorizing of emotion-behavior relations, as well as for practical considerations. In addition, a diversity workshop conducted with future diplomats showed indirect effects on solidarity-based collective action via diversity perception and superordinate group identification, thereby extending past research by including action intentions and going beyond the focus on grassroot-initiatives by presenting an implementable intervention for future leaders in a real world context.
Taken together, this dissertation provides important insights for the development of socio-psychological interventions. By integrating a diverse sample, including members of institutions on meso- and macro-levels (non-governmental organizations and future politicians) of our society, this dissertation presents a unique multi-perspective of host society members on the social conflict of refugee acceptance and support. Thereby, this work contributes to theoretical and practical advancement of how social psychology can contribute not only to negative peace – by for example (indirectly) reducing support of violence against refugees – but also to positive peace – by for example investigating precursors of hierarchy-challenging actions that enable equal rights.
Blockchain in Healthcare
(2020)
The underlying characteristics of blockchain can facilitate data provenance, data integrity, data security, and data management. It has the potential to transform the healthcare sector. Since the introduction of Bitcoin in the fintech industry, the blcockhain technology has been gaining a lot of traction and its purpose is not just limited to finance. This thesis highlights the inner workings of blockchain technology and its application areas with possible existing solutions. Blockchain could lay the path for a new revolution in conventional healthcare systems. We presented how individual sectors within the healthcare industry could use blockchain and what solution persists. Also, we have presented our own concept to improve the existing paper-based prescription management system which is based on Hyperledger framework. The results of this work suggest that healthcare can benefit from blockchain technology bringing in the new ways patients can be treated.
The distributed setting of RDF stores in the cloud poses many challenges. One such challenge is how the data placement on the compute nodes can be optimized to improve the query performance. To address this challenge, several evaluations in the literature have investigated the effects of existing data placement strategies on the query performance. A common drawback in theses evaluations is that it is unclear whether the observed behaviors were caused by the data placement strategies (if different RDF stores were evaluated as a whole) or reflect the behavior in distributed RDF stores (if cloud processing frameworks like Hadoop MapReduce are used for the evaluation). To overcome these limitations, this thesis develops a novel benchmarking methodology for data placement strategies that uses a data-placement-strategy-independent distributed RDF store to analyze the effect of the data placement strategies on query performance.
With this evaluation methodology the frequently used data placement strategies have been evaluated. This evaluation challenged the commonly held belief that data placement strategies that emphasize local computation, such as minimal edge-cut cover, lead to faster query executions. The results indicate that queries with a high workload may be executed faster on hash-based data placement strategies than on, e.g., minimal edge-cut covers. The analysis of the additional measurements indicates that vertical parallelization (i.e., a well-distributed workload) may be more important than horizontal containment (i.e., minimal data transport) for efficient query processing.
Moreover, to find a data placement strategy with a high vertical parallelization, the thesis tests the hypothesis that collocating small connected triple sets on the same compute node while balancing the amount of triples stored on the different compute nodes leads to a high vertical parallelization. Specifically, the thesis proposes two such data placement strategies. The first strategy called overpartitioned minimal edge-cut cover was found in the literature and the second strategy is the newly developed molecule hash cover. The evaluation revealed a balanced query workload and a high horizontal containment, which lead to a high vertical parallelization. As a result these strategies showed a better query performance than the frequently used data placement strategies.