Refine
Year of publication
- 2020 (76) (remove)
Document Type
- Doctoral Thesis (36)
- Part of Periodical (22)
- Bachelor Thesis (7)
- Master's Thesis (7)
- Article (1)
- Preprint (1)
- Report (1)
- Study Thesis (1)
Has Fulltext
- yes (76) (remove)
Keywords
- Datenschutz (2)
- Forschen in Koblenz (2)
- Junges Forschen (2)
- Persönlichkeit (2)
- Wissenschaft Koblenz (2)
- model-based (2)
- 3D Modell Referenz Bildsynthese (1)
- 3D-Vermessung (1)
- Abdrift (1)
- Abwasserreinigung (1)
Institute
- Institut für Management (14)
- Institut für Computervisualistik (10)
- Fachbereich 7 (9)
- Zentrale Einrichtungen (9)
- Fachbereich 8 (4)
- Institut für Integrierte Naturwissenschaften, Abt. Physik (4)
- Fachbereich 4 (3)
- Institut für Integrierte Naturwissenschaften, Abt. Biologie (3)
- Institut für Umweltwissenschaften (3)
- Institut für Kulturwissenschaft (2)
Social media platforms such as Twitter or Reddit allow users almost unrestricted access to publish their opinions on recent events or discuss trending topics. While the majority of users approach these platforms innocently, some groups have set their mind on spreading misinformation and influencing or manipulating public opinion. These groups disguise as native users from various countries to spread frequently manufactured articles, strong polarizing opinions in the political spectrum and possibly become providers of hate-speech or extremely political positions. This thesis aims to implement an AutoML pipeline for identifying second language speakers from English social media texts. We investigate style differences of text in different topics and across the platforms Reddit and Twitter, and analyse linguistic features. We employ feature-based models with datasets from Reddit, which include mostly English conversation from European users, and Twitter, which was newly created by collecting English tweets from selected trending topics in different countries. The pipeline classifies language family, native language and origin (Native or non-Native English speakers) of a given textual input. We evaluate the resulting classifications by comparing prediction accuracy, precision and F1 scores of our classification pipeline to traditional machine learning processes. Lastly, we compare the results from each dataset and find differences in language use for topics and platforms. We obtained high prediction accuracy for all categories on the Twitter dataset and observed high variance in features such as average text length especially for Balto-Slavic countries.
The maintenance strategy “predictive maintenance”, which is characterized by predicting the failure behavior of technical units based on modern sensor technology, plays a key role in smart factories against the background of an industry 4.0. This paper contains an evaluation of the current state of research on this strategy and gives an overview of the areas of application to date. With the aid of a qualitative video analysis, the implementation in the industries and company divisions involved and the type of goods monitored are examined. The analyzed video clips were uploaded to YouTube for example for marketing purposes by various companies with different perspectives on predictive maintenance. The video analysis was realized by applying a previously defined coding plan to the video material. The results show a predominant application in the manufacturing industry, in which predictive maintenance is used to monitor plants and machines. In addition, the strategy is also mainly applied to means of transport used for freight and passenger transport in various infrastructures. As a result of the video analysis, the currently high need for explanation of predictive maintenance becomes visible. By looking at these explanations, one also learns something about the special features that distinguish it from other maintenance strategies.
Eine zutreffende Diagnose über den aktuellen Kenntnisstand der jeweiligen Schülerinnen und Schüler ist notwendig, um adäquat in Gruppenarbeitsprozesse intervenieren zu können. Von diesem Zusammenhang wird in der Literatur weit-gehend ausgegangen, jedoch gibt es bisher kaum empirische Studien, die diesen belegen. Die vorliegende Arbeit widmet sich schwerpunktmäßig dem Interventi-onsverhalten von Studierenden. Dabei wird die prozessdiagnostische Fähigkeit „Deuten“ zugrundegelegt, um unterschiedliches Interventionsverhalten auf diese Fähigkeit zurückführen zu können. Sowohl beim Aufbau diagnostischer Fähig-keiten als auch bei der (Weiter-)Entwicklung des eigenen Lehrerhandelns gilt Reflexion als hilfreich. Entsprechend wird auch das Zusammenspiel von Pro-zessdiagnose und Reflexionsverhalten sowie von Interventionsverhalten und Reflexionsverhalten untersucht.
Für die Erhebung der prozessdiagnostischen Fähigkeit „Deuten“ wurden drei Videovignetten erstellt und in das Videodiagnosetool ViviAn eingebunden. Die Videovignetten zeigen jeweils vier Schülerinnen, die sich mit dem Thema „Ter-me“ beschäftigen. Im Rahmen eines Lehr-Lern-Labores wurden über vier Se-mester hinweg alle teilnehmenden Studierenden dazu angehalten, die Videovig-netten zu bearbeiten. Ebenso konzipierten sie jeweils zu dritt eine Laborstation im Mathematik-Labor „Mathe ist mehr“ und erprobten diese mit einer Schul-klasse. Dabei wurden die Interventionen der Studierenden in die Gruppenarbeits-prozesse der Schülerinnen und Schüler videographiert. Anschließend reflektierten die Studierenden in Kleingruppen über die Erprobungen und über die getätigten Interventionen. Die Reflexionsgespräche wurden ebenfalls videographiert.
Es zeigt sich, dass die Studierenden, die sich zum Zeitpunkt der Erhebung im Masterstudium befanden, noch Entwicklungsspielraum in Bezug auf ihre pro-zessdiagnostische Fähigkeit „Deuten“ besitzen. Im Hinblick auf die Interventio-nen waren responsive Interventionen häufiger angemessen als invasive Interven-tionen, wobei responsive Internvetionen auch vergleichsweise häufiger dazu führten, dass mehr Schülerinnen und Schüler nach der Intervention aktiv waren. Studierende mit höherer prozessdiagnostischer Fähigkeit „Deuten“ intervenierten jedoch häufiger invasiv und tätigten dabei trotzdem angemessenere und aktivie-rendere Interventionen als ihre Kommilitoninnen und Kommilitonen. Entspre-chend scheint sich die prozessdiagnostische Fähigkeit „Deuten“ positiv auf die Interventionen der Studierenden auszuwirken und sollte daher bereits im Rah-men des (Lehramts-)Studiums verstärkt geschult werden.
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.
Die Forschungsarbeit beschäftigt sich mit der zentralen Frage, welche Erfolgsfaktoren einen Effekt auf eine erfolgreiche Crowdfunding Kampagne haben. Als Untersuchungsfeld stehen deutsche Startup-Unternehmen im Fokus, die bereits erfolgreiche CrowdfundingKampagnen durchgeführt haben.
Zur Beantwortung dieser Frage wird zunächst eine systematische Literaturanalyse durchgeführt, durch die relevante Erfolgsfaktoren für eine Crowdfunding-Kampagne ermittelt werden. Diese Faktoren werden anschließend einem Mixed-Method-Ansatz unterzogen, bei dem qualitative Ergebnisse, basierend auf einer Fallstudienforschung, mit den statistisch ausgewerteten quantitativen Ergebnissen aus der Fragebogenforschung verglichen und überprüft werden. Dabei liegt der Fokus insbesondere auf der Identifikation von signifikanten Wirkungszusammenhängen zwischen den Erfolgsfaktoren und einer erfolgreichen Crowdfunding-Kampagne.
Im Ergebnis konnten diesbezüglich Wirkungszusammenhänge innerhalb dieser Thesis nachgewiesen werden. Sechs der festgestellten Zusammenhänge aus der qualitativen Analyse werden durch die quantitative Analyse bestätigt. Es konnte jedoch keine Signifikanz der Korrelationen festgestellt werden. Auch die Hypothese, dass sich die Erfolgsaussichten einer Kampagne durch eine höhere Anzahl jeweils kombinierter Erfolgsfaktoren erhöhen, wurde sowohl in der qualitativen als auch in der quantitativen Betrachtung widerlegt. Demnach galt es für den Autor der vorliegenden Thesis künftige Forschungsfelder zu definieren, die die ermittelten Ergebnisse erweitern und konkretisieren. Es bleibt beispielsweise einer weiterführenden Forschung überlassen, zu klären, ob bestimmte Kombinationen der Erfolgsfaktoren zu signifikanten Wirkungszusammenhängen führen. Darüber hinaus bietet sich eine weiterführende statistische Regressionsanalyse an, um die kausalen Effekte zu untersuchen und Prognosen für erfolgreiche Crowdfunding-Kampagnen zu formulieren.
The scientific interest in nonsuicidal self-injury (NSSI) has increased in the last two decades. High prevalence and comorbidity rates, low quality of life and increased risk of suicidality highlight the importance of this research field. The present thesis focuses on intra- and interpersonal factors associated with the development and maintenance of NSSI.
The aim of study 1 was the examination of personality traits of adolescents with NSSI without Borderline Personality Disorder (NSSI-BPD), adolescents with NSSI and BPD (NSSI+BPD), clinical controls (CC) and nonclinical controls (NC). Results showed that adolescents with NSSI disorder scored significantly higher on novelty seeking and harm avoidance and lower on persistence, self-directedness, and cooperativeness than CC. In adolescents with NSSI+BPD this personality pattern was even more pronounced than in adolescents with NSSI-BPD.
Adolescents´ NSSI leads to distress that affects the whole family system, often resulting in conflicts and disrupted family communication and functioning. Parents report feelings of distress, insecurity and helplessness. Adolescents with NSSI report more parental criticism and control and less support than adolescents without NSSI. Study 2 investigated the parenting behavior in families of adolescents with NSSI. Adolescents with NSSI reported less maternal warmth and support than NC adolescents. Mothers of adolescents with NSSI showed higher psychopathology scores than NC mothers and less parental satisfaction than CC and NC mothers.
Siblings are also reported to suffer from changes in family dynamics. The aim of study 3 was to examine the sibling relationship quality of adolescents with NSSI, CC and NC. Siblings reported a wide range of negative emotional and familial consequences as a result of their sister´s NSSI. Siblings of adolescents with NSSI experienced significantly more coercion in the relationship with their sister compared to CC and NC siblings. Adolescents with NSSI reported significantly less warmth and empathy in the sibling relationship and higher rivalry scores between their siblings and themselves than NC adolescents. For both, adolescents with NSSI and their siblings, associations were found between sibling relationship quality and internalizing problems.
Study 4 aimed to further explore the family emotional climate. Therefore, the level of expressed emotion (EE) was assessed in adolescents with NSSI, CC, NC and their mothers. Parental high EE (HEE) is linked to adolescent NSSI, especially parental criticism seems to be strongly associated with NSSI. Previous research into NSSI and EE has focused on parental EE, however, the conceptualization of EE as a unidirectional construct from parent to child may present an incomplete picture. Therefore, the current study included both, adolescent and maternal EE. Adolescents in the NSSI and CC group more often met criteria for HEE than NC. Adolescents with NSSI exhibited significantly more covert criticism and critical tone toward their mothers than CC and NC. HEE of adolescents with NSSI was associated with a range of difficulties in emotion regulation. For the total sample, moderate concordance was found between adolescents and mothers EE-status.
The research presented in this thesis has important clinical implications. The differences in personality traits of adolescents with NSSI with and without BPD underline the need for a dimensional personality assessment as well as specific treatment programs for adolescents with NSSI-BPD. Problems within the family are frequent triggers for NSSI. Therefore, interventions for adolescents with NSSI should include both, the improvement of emotion regulation and family interaction and communication. Along with the reduction of negative relationship aspects, psychotherapy should also focus on the enhancement of positive relationship quality. The emotional burden of family members stresses the need for emotional and practical support for parents and siblings.
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.
Perfluorocarboxylic acids (PFCA) are substances of anthropogenic origin and have been used for several decades. These compounds are a new class of environmental pollutants. Their high surface activity, thermal stability, amphipathicity and weak intermolecular interactions lead to persistence and bioaccumulation. Therefore, there is a great need for reliable analytical methods for detecting the presence and determination of concentration in both environmental samples and everyday products. GC-MS is a cost-effective alternative and supplement to established LC-MS/MS methods. The greatest challenge in the method development is the derivatization reaction. Many of the previously published derivatization reactions for PFCA are time consuming and require high reaction temperatures or toxic reagents.
In the present dissertation, two new derivatization reactions for PFCA have been developed and optimized. The first part of the thesis shows the development and optimization of the reaction with triethylsilanol in water. In addition to optimizing the reaction, classical solid-phase extraction was modified to simplify the sample preparation.
In the second part of the work, the reaction products of perfluorooctanoic acid (PFOA) with N,N-dimethylformamide dimethyl acetal (DMF-DMA) and -diethyl acetal (DMF-DEA) were identified. From these measurements, it follows that both DMF-DMA and DMF-DEA in the presence of PFOA form an iminium cation, which leads to salt formation. This PFOA salt react further in the GC injector and a corresponding amine is produced.
In the last part of the thesis, an analytical method based on the DMF-DMA reaction was developed. The matrix effects have been described in detail. The method has been successfully applied for three different types of samples: dental floss, textiles and sewage sludge. The results were verified by LC-MS/MS analysis in an external laboratory. The differences between the PFCA values for a spiked sample measured by GC-MS and LC-MS/MS were less than 10%.