The proliferation of online abuse on social media platforms has emerged as a significant concern, negatively impacting users' mental health and online experiences. While the Natural Language Processing (NLP) community has developed various computational methods for abuse detection, including Support Vector Machines (SVMs) and Convolutional Neural Networks (CNNs), existing approaches predominantly focus on identifying explicit forms of abuse. This narrow focus overlooks subtle and contextual forms of online harassment, which can be equally damaging to users' wellbeing. This thesis presents a novel approach to online abuse detection by integrating contextual embeddings with sentiment analysis features through the fine-tuning of Large Language Models (LLMs). Our methodology leverages a comprehensive dataset of 47,000 annotated tweets for training, combined with sentiment analysis capabilities developed using 50,000 IMDB movie reviews. The system employs DistilBERT architecture to develop a sophisticated detection framework capable of identifying six distinct categories of abuse: ethnicity-based, age-based, gender-based, religion-based, other cyberbullying, and non-cyberbullying content. The author established a rigorous evaluation framework employing multiple metrics, including accuracy, recall, and F1 score, to assess the model's performance in detecting both explicit and nuanced forms of online abuse. The integrated system achieved an overall accuracy of 85\% across 6 categories on the cyberbullying dataset, outperforming other methodologies applied to the same data. In direct comparison, our approach— which uniquely combines contextual embeddings with sentiment analysis—demonstrated significant improvements over traditional fine-tuning methods, such as those using only BERT or RoBERTa, particularly in detecting subtle forms of abuse. Most notably, our system was more effective at identifying passive-aggressive content and context-dependent harassment, challenges that often cause conventional detection methods to fall short. This enhanced performance can be attributed to the model's ability to capture nuanced linguistic cues through its integrated analysis of both contextual information and sentiment, thereby offering a more refined interpretation of potentially harmful content. This research emphasizes the critical importance of incorporating subtle abuse detection into online content moderation systems. By developing more sophisticated detection methods that can identify both overt and nuanced forms of harassment, this work contributes to the creation of safer and more inclusive online spaces that facilitate constructive dialogue. The findings of this study have significant implications for the development of more effective content moderation tools and the broader goal of fostering healthier online communities.

Die vorliegende Arbeit untersucht die Forschungsfrage Welche fachspezifischen Faktoren beeinflussen den Einsatz digitaler Medien in der sportlichen Lehre von rheinland-pfälzischen Sportlehrkräften und Lehrkräftebildner:innen? Grundlage der Untersuchung bildet das Will-Skill-Tool-Modell zur Technologieintegration von Knezek et al. (2003), das drei zentrale Einflussfaktoren differenziert: (1) Die individuellen Haltungen und Einstellungen der Lehrperson gegenüber digitalen Medien (Will), (2) die subjektiv wahrgenommenen digitalisierungsbezogenen Kompetenzen (Skill) und (3) der Zugang zu technischer Infrastruktur (Tool). Im Rahmen der Studie wurden 55 problemzentrierte leitfadengestützte Interviews mit rheinland-pfälzischen Sportlehramtsstudierenden, Lehramtsanwärter:innen, Lehrkräften, Seminarleiter:innen des Unterrichtsfaches Sport und Hochschuldozierenden, welche in der universitären Ausbildung von Sportlehrkräften involviert sind, durchgeführt. Die Ergebnisse zeigen, dass die Beteiligten digitale Medien maßvoll einsetzen, wobei der Schwerpunkt auf der Nutzung von Videos und der Entlastung der Lehrperson liegt. Die Befragten stehen dem Einsatz digitaler Medien in der sportlichen Lehre überwiegend positiv gegenüber, empfinden jedoch häufig ein Spannungsfeld zwischen Mediennutzung und einem potenziellen Verlust an Bewegungszeit. Die erforderlichen digitalisierungsbezogenen Kompetenzen sollen aus Sicht der Be-fragten das Lernen mit Medien als Unterrichtswerkzeuge unterstützen, während kritisch und reflexive Aspekte kaum thematisiert werden. Im Vergleich zu Schulen wird die technische Ausstattung von Sporthallen als verbesserungswürdig wahrgenommen, was den Medieneinsatz hemmt. Die COVID-19-Pandemie hatte laut den Befragten einen signifikanten Einfluss auf den Medieneinsatz, die persönlichen Einstellungen und die individuellen digitalisierungsbezogenen Kompetenzen. Die Ergebnisse deuten darauf hin, dass der Medieneinsatz nicht mehr ausschließlich als abhängige Variable betrachtet werden kann, sondern selbst Einfluss auf die Aspekte Will und Skill ausübt. Die spezifische Betrachtung der Modellkategorien legt nahe, dass dem Aspekt Tool für die sportliche Lehre eine besondere Bedeutung zukommt. Handlungsempfehlungen zur Förderung eines nachhaltigen und effektiven Einsatzes digitaler Medien in der sportlichen Lehre umfassen die verbindliche Integration digitaler Medien in Fachcurricula, die Gestaltung niederschwelliger Fortbildungen, die adäquate Ausstattung von Sporthallen und den Ausbau von Unterstützungsstrukturen.

Population genetics investigates genetic diversity and its changes within and between populations over space and time. Genetic diversity is important for fitness, adaptive capacity, and the survival of populations and is influenced by several factors, such as mutation, selection, genetic drift and gene flow. Copper butterflies (Lycaena) are suitable for analysing structures influencing population connectivity as they potentially form more or less closed populations. However, very little is known about their genetic diversity and what influences it. Therefore, this thesis (1) provides newly developed microsatellite markers and uses genetic markers (2) to investigate genetic diversity across four different Lycaena species in the European Alps and to determine (3) which geographic and species specific factors influence population structure, (4) which large- and small-scale structures impact the population structure, (5) how natural and anthropogenic structures influence the population structure within an Alpine valley, and (6) whether and how genetic diversity changes over time. It was shown that the postglacial relict species L. helle has a relatively high genetic diversity compared to the other three species investigated. This suggests that L. helle is still able to adapt to environmental changes. Low genetic diversity was found in L. tityrus subalpinus, although high gene flow was found within one population of this species. High mountain ridges and large river valleys can act as dispersal barriers for Copper butterflies and thus have an impact on population structures. Here, dispersal ability as a species-specific factor also plays an important role, as some barriers are less likely to affect the population structure in the more mobile species L. virgaureae. Furthermore, forests, ravines and roads, but not small rivers, represent dispersal barriers for L. tityrus subalpinus within an Alpine valley. Finally, over ten years, the genetic diversity of L. hippothoe eurydame has decreased, whereas that of L. helle has remained stable. Against the backdrop of increasing global changes, it is important to understand the genomic underpinning of population structure and adaptation as well as to investigate and monitor whether populations are able to adapt to changing environmental conditions.

During the development phase of plastic components, simulations are being used to an increasing extent. Against the background of product requirements and the inevitable necessity of conserving resources, the expanded use of simulation tools is an essential part of the solution. Among available methods, but so far underutilized with respect to real-life processes, is the molecular dynamics simulation. By the use of this method it is possible to visualize the physical processes occurring on the microscopic level, as e.g. those that arise during plastics processing. This thesis examines how boundary conditions, which mimic the extrusion blow molding process, affect the behavior of polyethylene on the microscopic level. A mesoscopic model (coarse-graining) is applied to describe the polymer. Initially, this model is verified by determining material properties. The uniaxial tensile test is modeled on the micro-scale to identify parameters such as the elastic modulus, yield stress, and Poisson’s ratio. Additionally, thermal properties, particularly those characterizing the crystallization behavior, are identified. The objective of these investigations is the microscopic observation and quantification of effects that occur during dynamic stretching and crystallization processes. The calculated properties show good agreement with the experimental data, especially regarding the thermal parameters. Qualitatively, the stress-strain behavior is reproduced in alignment with experimentally observed results. However, the short time scale of the simulation models leads to micromechanical behavior that is more extreme than what is monitored on a macroscopic level. By extending the simulation models, biaxial stretching processes are simulated. These stretching processes resemble the situation during the inflation of the parison in the extrusion blow molding process. The examination of various cooling conditions, particularly by the use of mold constraints, is another focus of the investigations. The analysis of the biaxially stretched simulations reveals that disentanglement processes during stretching dominate the further development of polymer systems. It is possible to quantify the dynamics of crystallization processes depending on the degree of stretching and cooling conditions through various parameters (distribution of entanglement points, local orientations). The results indicate that coarse-grained molecular dynamics simulations are able to significantly enhance the micromechanical understanding of local events occurring during plastic processing.

Entrepreneurship and entrepreneurship education have expanded together, and their conceptual and methodological challenges do not prevent the implementation of entrepreneurial education in educational contexts. The desire for a global workforce that can handle uncertainties and solve problems that cannot be solved by pure analytical inquiry drives the rapidly expanding number of educational programs and activities that are design-based. A growing number of educational programs for entrepreneurs increasingly incorporate design-based methods. However, design thinking-based theoretical assumptions may also be lacking. Despite growing academic interest in design thinking and entrepreneurship education, little is known about design thinking in higher entrepreneurship education, especially in Eastern nations. A Western teaching method, entrepreneurial design thinking may be adapted to many cultures. In this instance, the West has established entrepreneurship education as a respectable study subject and teaching practice in higher education over the past 40 years. The Eastern nations' occurrence varies, including Indonesia. Indonesia is an intriguing research subject since it has over 50% youth due to its abundant natural resources. However, it needs more opportunity-based entrepreneurs and requires assistance in implementing entrepreneurship education with a more innovative, design-based, and successful method. Entrepreneurial design thinking fulfills this demand. Indonesian students and teachers' norm-based attitudes and cultural mindsets towards a new western creative method may hinder entrepreneurial design thinking's acceptance. The literature review found that Indonesian university students are collaborative, compassionate, and practical, like design thinkers. However, they may also be risk-averse, self-restrained, and dependent on teachers as stereotypical Asians. Classroom space, educators' design thinking competence, and university or institution support are further barriers. Additional study into these challenges is needed to adapt design thinking to Indonesian entrepreneurial higher education culturally. Based on the above research needs, the purpose of this research endeavor is to look into the cultural nuances of the design thinking technique for entrepreneurial higher education and postulate how it could be adapted to other cultures, especially in Indonesia. This thesis uses deductive and qualitative case-study research methods. In particular, the latter used thematic analysis (Braun & Clarke, 2006; Terry & Hayfield, 2021) as the data analysis technique to provide a means and tools for understanding from interviews, class observation, and literature studies. Since this thesis follows the constructivist-relativist research paradigm, it explores contextual and cultural differences in Indonesian entrepreneurship education and its potential and obstacles to adapt the Western teaching methodology of entrepreneurial design thinking in higher education. In summary, this study searches for elements that might aid or hinder the cross-cultural adaption of entrepreneurial design thinking. This research wants to understand how cross-cultural adaptability fits into entrepreneurial design thinking research, especially for Indonesia. This thesis aims to provide new theoretical insights and practical advice on adapting entrepreneurial design thinking from Western to Eastern cultures. From the findings, this thesis concluded at least seven educational value differentials before adaptation from the exhaustive literature and case study evaluations. For Indonesian entrepreneurship higher education institutions to use entrepreneurial design thinking, they must consider educational culture, technological infrastructure, language, primary audience, learning and teaching style, reasoning patterns, and social-cultural environment. This study provides four practical adaptation recommendations: socialization, externalization, combination, and internalization. Finally, this research demonstrated that cross-cultural adaption of entrepreneurial design thinking in Indonesia might be difficult but worthwhile. This thesis' case study, "School of Business Management – Bandung Institute of Technology (SBM ITB)", showed that Indonesian entrepreneurial higher education might use design thinking as a teaching approach. All stakeholders must improve internally and publicly. Thus, this study recommended integrating most Indonesian higher education institutions' entrepreneurship teaching approaches with a "student-centered" approach that stresses business mentorship, uses design thinking tools and processes, and links them to students' entrepreneurial initiatives. To summarize, this research contributes to the field since it draws on and combines the findings of several other fields of study, including entrepreneurial education, design thinking, and cross-cultural adaptation. This study stepped out of the "usual and proper" pedagogical ruts to investigate "non-human" cross-cultural adaptability. It has attempted to apply these ideas to a real-world, unique case study in a developing nation (in this case, Indonesia).