1582 research outputs found
Sort by
Simulation of emissions from pottery kilns in the Roman period
Full text linkPresent-day air quality is known through dense monitoring and extensive pollu-
tion control mechanisms. In contrast, knowledge of historical pollution,
particularly before the industrial revolution, is accessible only through occasional
reports of singular local events and through natural archives such as ice or
sediment cores that record global-scale pollution. However, the regular local to
regional pollution that most affects human life is hardly known. Historical
sciences have argued both for and against significant air pollution in and around
historic cities and manufacturing sites. For the Roman era, it has been
hypothesized that air quality played a role in several patterns of action of the period.
However, to the author's knowledge, there are no quantitative studies of
Roman emissions. Using the results of modern experimental archaeology, this
study attempts to quantify the emissions from Roman pottery kilns and their
impact on surrounding human settlements. It is shown that although the
pollution did not reach today's limits, it must have approached levels known to cause
adverse health effects. A series of additional test simulations have been
conducted to determine how these first results might be improved in the future
Under pressure: moods, affects and the violence of everyday life in a Spanish migrant detention centre
Full text linkWhat does it mean when the future of one’s life is exposed to the inscrutable will of an intangible other? And what are the possibilities of still asserting oneself when pushed to the limit? Nuancing the feelings of different actors in a detention centre and analysing how everyday moods, affects and violence intertwine, I explore how the randomly cruel and often-inexplicable logic of the contemporary deportation regime pushes migrants to their limits. Taking as my starting point the argument that deportation practices are effective because they operate on an affective level, I show how affective experiences manifest themselves bodily and how violent practices and discourses reverberate in bodies. I argue that ‘bodies under pressure’ are testimonies of racialised histories of exclusion, and I show how they become calls for social recognition. Exploring small, often-unintended acts of rebellion against exhausting deportation practices, I stress the existential necessity and social importance of including oneself in the realm of meaning
Knowledge Acquisition and Transfer – Three Meta-Analytic Investigations
Full text linkKnowledge acquisition comprises various processes. Each of those has its dedicated research domain. Two examples are the relations between knowledge types and the influences of person-related variables. Furthermore, the transfer of knowledge is another crucial domain in educational research. I investigated these three processes through secondary analyses in this dissertation. Secondary analyses comply with the broadness of each field and yield the possibility of more general interpretations. The dissertation includes three meta-analyses: The first meta-analysis reports findings on the predictive relations between conceptual and procedural knowledge in mathematics in a cross-lagged panel model. The second meta-analysis focuses on the mediating effects of motivational constructs on the relationship between prior knowledge and knowledge after learning. The third meta-analysis deals with the effect of instructional methods in transfer interventions on knowledge transfer in school students. These three studies provide insights into the determinants and processes of knowledge acquisition and transfer. Knowledge types are interrelated; motivation mediates the relation between prior and later knowledge, and interventions influence knowledge transfer. The results are discussed by examining six key insights that build upon the three studies. Additionally, practical implications, as well as methodological and content-related ideas for further research, are provided.Wissenserwerb umfasst verschiedene Prozesse. Jedem dieser Prozesse ist ein eigener Forschungsbereich gewidmet. Zwei Beispiele sind die Zusammenhänge zwischen Wissenstypen und die Einflüsse personenbezogener Variablen. Darüber hinaus ist der Transfer von Wissen ein weiterer wichtiger Bereich in der Bildungsforschung. In dieser Dissertation werden diese drei Prozesse mithilfe von Sekundäranalysen untersucht. Sekundäranalysen werden der Breite des jeweiligen Feldes gerecht und bieten die Möglichkeit allgemeinerer Interpretationen. Die Dissertation umfasst drei Meta-Analysen: Die erste Meta-Analyse berichtet über Befunde zu prädiktiven Beziehungen zwischen konzeptuellem und prozeduralem Wissen in Mathematik in Form eines Cross-Lagged-Panel-Modells. Die zweite Meta-Analyse befasst sich mit den vermittelnden Effekten motivationaler Konstrukte in der Beziehung zwischen Vorwissen und Wissen nach dem Lernen. Die dritte Meta-Analyse untersucht die Wirkung von Lehrmethoden in Transferinterventionen auf den Wissenstransfer bei Schüler*innen. Diese drei Studien bieten Einblicke in die Determinanten und Prozesse des Wissenserwerbs und -transfers. Wissenstypen sind miteinander verknüpft, Motivation vermittelt die Beziehung zwischen früherem und späterem Wissen und Interventionen beeinflussen den Wissenstransfer. Die Ergebnisse werden anhand von sechs Schlüsselerkenntnissen diskutiert, die auf den drei Studien aufbauen. Darüber hinaus werden praktische Implikationen sowie methodische und inhaltliche Anregungen für weitere Forschung gegeben
Les conflits scolaires comme opportunité d’apprentissage de la démocratie
Full text linkLes conflits sont inévitables. Ils surviennent même au sein d’établissements scolaires empreints d’une culture démocratique. L’instauration d’une culture d’échange constructif sur les questions conflictuelles permet toutefois d’utiliser les litiges comme des opportunités d’apprentissage et de développement personnel et organisationnel
Mixed-Integer Optimization for Semi-Supervised Learning with Cardinality Constraints: Support Vector Machines, Classification Trees, and Random Forests
Full text linkIn machine learning, classification is the task of predicting a label for each point within a data set. When the class of each point in the labeled subset is already known, this information is used to recognize patterns and make predictions about the points in the remainder of the set, referred to as the unlabeled set. This scenario falls in the field of supervised learning.
However, the number of labeled points can be restricted, because, e.g., it is expensive to obtain this information. Besides, this subset may be biased, such as in the case of self-selection in a survey. Consequently, the classification performance for unlabeled points may be limited. To improve the reliability of the results, semi-supervised learning tackles the setting of labeled and unlabeled data. Moreover, in many cases, additional information about the size of each class can be available from undisclosed sources.
This cumulative thesis presents different studies to combine this external cardinality constraint information within three important algorithms for binary classification in the supervised context: support vector machines (SVM), classification trees, and random forests. From a mathematical point of view, we focus on mixed-integer programming (MIP) models for semi-supervised approaches that consider a cardinality constraint for each class for each algorithm.
Furthermore, since the proposed MIP models are computationally challenging, we also present techniques that simplify the process of solving these problems. In the SVM setting, we introduce a re-clustering method and further computational techniques to reduce the computational cost. In the context of classification trees, we provide correct values for certain bounds that play a crucial role for the solver performance. For the random forest model, we develop preprocessing techniques and an intuitive branching rule to reduce the solution time. For all three methods, our numerical results show that our approaches have better statistical performances for biased samples than the standard approach
Russische Spionage und Sabotage in Deutschland – Hybride Kriegsführung ohne Kriegserklärung
Full text link
The Collection of Eighteenth-Century French Novels 1751–1800
Full text linkThe French Enlightenment is a pivotal period in European intellectual and literary history, which can be studied through this dataset of French novels first published between 1751 and 1800. This collection contains 200 French novels in TEI/XML, encoded according to the ‘level-1 schema’ of the European Literary Text Collection (ELTeC), and carefully compiled to reflect the known historical publication of French Novels in that period regarding publication year, gender of author and narrative form. The dataset is connected to a bigger knowledge graph of 331,671 Resource Description Framework triples (RDF) built within the project ‘Mining and Modeling Text’ at Trier University, Germany (2019–2023)
Energy stored in soil organic matter is influenced by litter quality and the degree of transformation – A combustion calorimetry study
Full text linkThe turnover and stabilization of organic matter (OM) in soils depend on mass and energy fluxes. Understanding the energy content of soil organic matter (SOM) is therefore of crucial importance, but this has hardly been studied so far, especially in mineral soils. In this study, combustion calorimetry (bomb calorimetry) was applied to determine the energy content (combustion enthalpy, ΔCH) of various materials: litter inputs, forest floor layers (OL, OF, OH), and bulk soil and particulate organic matter (POM) from topsoils (0–5 cm). Samples were taken from 35-year-old monocultural stands of Douglas fir (Pseudotsuga menziesii), black pine (Pinus nigra), European beech (Fagus sylvatica), and red oak (Quercus rubra) grown under highly similar soil, landscape and boundary conditions. This allowed to investigate the influence of the degree of transformation and litter quality on the ΔCH of SOM. Tree species fuel the soil C cycle with high-energy litter (38.9 ± 1.1 kJ g−1C) and fine root biomass (35.9 ± 1.1 kJ g−1C). As plant material is transformed to SOM, ΔCH decreases in the order: OL (36.8 ± 1.6 kJ g−1C) ≥ OF (35.9 ± 3.7 kJ g−1C) > OH (30.6 ± 7.0 kJ g−1C) > 0–5 cm bulk soil (22.9 ± 8.2 kJ g−1C). It indicates that the energy content of OM decreases with transformation and stabilization, as microorganisms extract energy from organic compounds for growth and maintenance, resulting in lower-energy bulk SOM. The POM fraction has 1.6-fold higher ΔCH compared to the bulk SOM. Tree species significantly affect ΔCH of SOM in the mineral soil with the lowest values under beech (12.7 ± 3.4 kJ g−1C). The energy contents corresponded to stoichiometric and isotopic parameters as proxies for the degree of transformation. In conclusion, litter quality, in terms of elemental composition and energy content, defines the pathway and degree of the energy-driven microbially mediated transformation and stabilization of SOM