Departing from Roland Barthes’ association of text and textile, and feminist theory on weaving as text production, this article analyzes the textile qualities of Inger Christensen’s “Letter in April” (1979) and Amalie Smith’s “Thread Ripper” (2020). In “Letter in April”, Christensen establishes a connection between writing and spinning or weaving through their shared temporality of varied repetition. In “Thread Ripper” Smith alludes to Christensen and makes of the continuity between text and textile not only the main theme of the book, but also its structuring principle. Through a materialist conception of the text, regarding it as a woven fabric, the article focuses on the textual patterns of the two works (stylistic figures in Christensen, graphic composition in Smith). The connection from Christensen to Smith leads to a further connection to ecocritical conceptions of weaving as no less than a cosmological principle. On a concluding note, the article argues that weaving is not only connecting, but also disconnecting, cutting.

Der vorliegende Beitrag greift die öffentliche Diskussion um den rechtspolitischen Umgang mit Hass, Hetze und Antisemitismus auf, die insbesondere nach dem Terroranschlag der Hamas am 07.10.2023 an Intensität und Dringlichkeit zugenommen hat. Dabei beleuchtet er einerseits das Straf- und Zivilrecht, legt andererseits einen besonderen Fokus auf öffentlich-rechtliche Konstellationen. Auf jedem dieser Gebiete werden Schwächen und Potenziale des Rechts und der Rechtsprechung aufgezeigt, zugleich aber auch die Grenzen staatlicher Gewalt verdeutlicht. Denn letztlich handelt es sich um ein gesellschaftliches Problem, dem – trotz aller Notwendigkeit staatlichen Handelns – in erster Linie durch Information, und erst in zweiter Linie durch das Recht begegnet werden muss.

　

In 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.