Eldorado - Repository of the TU Dortmund
Resources for and from Research, Teaching and Studying
This is the institutional repository of the TU Dortmund. Ressources for Research, Study and Teaching are archived and made publicly available.
Communities in Eldorado
Select a community to browse its collections.
Recent Submissions
Amtliche Mitteilungen der Technischen Universität Dortmund Nr.27/2025
(Technische Universität Dortmund, 2025-10-10)
Component-based synthesis using combinatory logic, intersection types and predicates
(2025) Stahl, Christoph; Rehof, Jakob; Düdder, Boris
In dieser Arbeit wird eine neue Version des Combinatory Logic Synthesizer ((CL)S) vorgestellt, der Combinatory Logic Synthesizer with Predicates (CLSP) genannt wird. Dieses neue Framework verwendet Inhabitation in der kürzlich entwickelten Finite Combinatory Logic with Predicates (FCLP) anstelle der Finite Combinatory Logic (FCL), die in früheren Iterationen verwendet wird. FCLP bietet neue Spezifikationsmöglichkeiten durch die Hinzufügung von parametrisierten Typen, die es erlauben, von einer endlichen Menge von Literalen abzuhängen und durch Prädikate eingeschränkt zu werden. Der Fokus von CLSP unterscheidet sich von früheren Iterationen von (CL)S darin, dass er mit Schwerpunkt auf Ausführungsgeschwindigkeit und Benutzerfreundlichkeit entwickelt wurde, anstatt auf beweisbare Korrektheit. Sowohl die theoretischen Grundlagen des Frameworks als auch die Implementierung in Python werden diskutiert. Eine Evaluation
des Frameworks in Form von Benchmarks wird präsentiert, die verschiedene Modellierungstechniken des neuen Frameworks miteinander vergleicht, sowie die
Synthesegeschwindigkeit von CLSP mit früheren Iterationen von (CL)S. Letzteres zeigt eine Verbesserung der Laufzeit von bis zu 100-fach, verglichen mit der schnellsten (CL)S Version, in bestimmten Benchmarks. Weiterhin wird eine prototypische parallele Implementierung des Inhabitationsalgorithmus vorgestellt, der eine beinahe lineare Beschleunigung in der Anzahl der verwendeten CPU-Kerne zeigt. Abschließend werden zwei Anwendungen des Frameworks diskutiert, eins, zum Finden von Strategien in einem Fragment von LTL, und eins, im Kontext von Simulationsmodellen in Materialflusssystemen. Dies zeigt, wie das Framework sowohl in theoretischen und praktischen Anwendungen genutzt werden kann.
Extracting economic narratives using natural language processing
(2025) Lange, Kai-Robin; Jentsch, Carsten; Rahnenführer, Jörg
As our society has become increasingly digitized across the last decades, the need to analyze unstructured data sources has grown in many sectors. In particular, economic and political researchers have been interested in quantitatively uncovering information found in texts, as theoretical research in these fields points towards the great effects of narratives on economic decision making. While such economic narratives can follow a myriad of definitions, they can generally be described as sense-making stories that can influence the reader's economic or political decisions in the future. Developing quantitative methods that can extract and analyze narratives from texts is therefore a major step to better understand market behavior or the decisions of policy makers, among others.
In this cumulative dissertation, I outline the research I have conducted concerning economic and political narratives with a particular focus on using diachronic language modeling that enables me to analyze narratives not in a vacuum, but rather observe narrative shifts over time. To do this, I first properly define economic narratives and provide an overview of language models I used in my research. I then proceed to summarize the methodology and contributions to the current research proposed in my papers, starting with works that do not consider a temporal component when extracting narratives from texts. These works show the development of narrative extraction techniques over time, starting from an unsupervised text classification method and a large pipeline of models specifically designed to handle the task, and ending with the use of state-of-the-art Large Language Models to extract narratives utilizing their great language understanding capabilities.
After covering atemporal narrative extraction methods, I focus on my works that utilize diachronic language modeling. I present two diachronic change detection methods, designed to identify points in time at which we can suspect a narrative shift. The first method uses the frequency of mentions of an entity in the media over time to detect a change, enabling an analysis of narratives surrounding that entity. The second method detects changes in the topics of the topic model LDA, allowing for an analysis of the corpus at large rather than a specific entity. I then propose a method to combine temporal and atemporal methods, with atemporal methods extracting narratives (or narrative shifts) and the detected change points. I further present software publications that contain all diachronic methods proposed in my works. Lastly, I conclude by giving an outlook on future research.
Model diagnostics and inference for count time series
(2025) Faymonville, Maxime; Jentsch, Carsten; Weiß, Christian
Count time series naturally arise when counting occurrences of things and events over time resulting in numerous applications in various fields. However, research on count time series and, more generally, discrete time series is not as advanced as research on classical continuous time series. This gap highlights the importance of developing appropriate methodologies for the discrete case in order to effectively address its unique characteristics.
This cumulative dissertation is based on five articles that collectively extend the research on count time series and discrete-valued time series in general. We provide an improved semi-parametric estimation procedure for the integer-valued autoregressive (INAR) model and develop an R package allowing for simulation, estimation and bootstrapping of INAR data. Furthermore, we present a semi-parametric INAR bootstrap procedure and prove its joint consistency for the estimation of the INAR coefficient and the innovation distribution. Finally, we propose a goodness-of-fit test on the whole INAR model class and provide methodology to conduct prediction in the setup of discrete time series in general. In addition to outlining these methodologies, we always validate our findings by extensive simulations and apply them on real-data examples. While three articles are published in
peer-reviewed journals, the other two are on arXiv and attached in their current version.
Simple Martingale Betting versus Labouchere Betting at Roulette: Simulations and Statistical Evaluations
(2023-05-23) Pflaumer, Peter
This paper investigates Labouchere betting systems through extensive simulations and a comprehensive comparison with martingale betting strategies. The analysis focuses on the statistical profit distribution of a Labouchere round, revealing it as a mixed random variable where the continuous part is approximated by the Gumbel-Gompertz distribution. The intricate nature of Labouchere's profit distribution is examined through simulations and statistical modeling, providing insights into its characteristics and significance. The study extends to practical considerations, presenting simulated outcomes under unlimited stakes,
further enriching the understanding of Labouchere betting dynamics. The paper contributes to both theoretical and applied aspects of Labouchere betting, shedding light on its complexities and implications for players and researchers alike.