Open Access

Social media has become an integral part of numerous individuals as well as organizations, with many services being used frequently by a majority of people. Along with its widespread use, the amount of information explodes when people use these services. This demands for efficient tools as well as methods to assist data management and retrieval. Annotating resources by keywords, known as the tagging task, is a solution to improve categorizability and findability of resources. However, tagging is a human, time-consuming task, which requires the user's focus to figure out many keywords in a short moment and manually enter them into the system. To encourage users to tag their resources more correctly and frequently, tag recommendation is adopted into the social tagging systems to suggest relevant keywords for resources. In this thesis, we will address the problem of personalized tag recommendation for images and present ways to solve this problem by combining the advantages of the user relation with the images' content. In order to suggest tags for unobserved images, their visual contents are used to replace the index-based information of the image entity in the tagging relations. Because the limitation of low-level features does not show the "content" of images, we propose to utilize a deep learning based approach to learn high-level visual features concurrently with the scoring-tag estimator. For the tag predictor, a latent factor model or a multi-layer perceptron is selected to compute scores of tags by which the top selected tags are sorted in descending order. As a further development upon our findings, we examine the inside and outside context of images to enhance the accuracy of estimators. Regarding the image-inside context, we are motivated by the fact that objects, such as cars or cats are influential on the user's selection criteria. Regarding the image-outside context, the image's surrounding text contributes to the clarity of the image's content for different users. We consider these contextual features as a supporting part which is combined with the mainly visual representation to enhance the tag recommendation performance. Finally, as an additional technique, transfer learning is also adapted to support the proposed models to overcome the limitations of too small training data and boost up their performance. This thesis demonstrates the usefulness and versatility of deep learning approaches for tag recommendation and highlights the importance of the learned image's content in predicting personalized tags. Directions for future work include semantic enhancements to context-based representation and extensions of the content-aware approaches to different recommendation scenarios.

In der vorliegenden Arbeit wird eine Unterrichtseinheit zum Themenbereich Graphentheorie im Mathematikunterricht der Grundschule durchgeführt und der Einfluss dieser Einheit auf die Motivation, das Selbstkonzept, die Einstellung zum Fach Mathematik und die mathematische Leistung der Schülerinnen und Schüler untersucht. Zu Beginn wird auf die Graphentheorie, ihre theoretischen Grundlagen und ihr didaktisches Potenzial eingegangen, um Gründe für den Einsatz dieses Themengebietes aufzuzeigen. Durch die Darstellung der Graphentheorie als ein mögliches Thema für den Mathematikunterricht der Grundschule werden psychologische Konstrukte herausgearbeitet, die sich mit den Unterrichtsinhalten verbessern lassen könnten. Hierbei handelt es sich um die Verstärkung der Motivation, des Selbstkonzepts, der positiven Einstellung zum Fach Mathematik und die Verbesserung der mathematischen Leistung. Anschließend werden die Stunden der Unterrichtseinheit mit den angestrebten Lernzielen und Kompetenzen dargelegt. Im Rahmen einer Interventionsstudie erhalten 40 Schülerinnen und Schüler zusätzlich zum Mathematikunterricht eine Unterrichtseinheit im Umfang von fünf Schulstunden zur Graphentheorie und es werden zu zwei Messzeitpunkten die Motivation, das Selbstkonzept, die Einstellung zum Fach Mathematik sowie die mathematische Leistung der Schülerinnen und Schüler erfasst. Die Auswirkungen dieser Unterrichtseinheit auf die genannten psychologischen Konstrukte werden anhand von t-Tests und Varianzanalysen untersucht. Die Ergebnisse der quantitativen Datenauswertung dieser Interventionsstudie zeigen signifikante Effekte mit kleinen und mittleren Effektstärken für einzelne Bereiche des Selbstkonzepts, der Einstellung sowie der mathematischen Leistung, sodass einige Hypothesen bestätigt werden können. Innerhalb der Interpretation und Evaluation dieser Effekte wird ein Ausblick auf anknüpfende Forschungen gegeben und es werden eigene Zielsetzungen dargelegt.

Automating machine learning by providing techniques that autonomously find the best algorithm, hyperparameter configuration and preprocessing is helpful for both researchers and practitioners. Therefore, it is not surprising that automated machine learning has become a very interesting field of research. Bayesian optimization has proven to be a very successful tool for automated machine learning. In the first part of the thesis we present different approaches to improve Bayesian optimization by means of transfer learning. We present three different ways of considering meta-knowledge in Bayesian optimization, i.e. search space pruning, initialization and transfer surrogate models. Finally, we present a general framework for Bayesian optimization combined with meta-learning and conduct a comparison among existing work on two different meta-data sets. A conclusion is that in particular the meta-target driven approaches provide better results. Choosing algorithm configurations based on the improvement on the meta-knowledge combined with the expected improvement yields best results. The second part of this thesis is more application-oriented. Bayesian optimization is applied to large data sets and used as a tool to participate in machine learning challenges. We compare its autonomous performance and its performance in combination with a human expert. At two ECML-PKDD Discovery Challenges, we are able to show that automated machine learning outperforms human machine learning experts. Finally, we present an approach that automates the process of creating an ensemble of several layers, different algorithms and hyperparameter configurations. These kinds of ensembles are jokingly called Frankenstein ensembles and proved their benefit on versatile data sets in many machine learning challenges. We compare our approach Automatic Frankensteining with the current state of the art for automated machine learning on 80 different data sets and can show that it outperforms them on the majority using the same training time. Furthermore, we compare Automatic Frankensteining on a large-scale data set to more than 3,500 machine learning expert teams and are able to outperform more than 3,000 of them within 12 CPU hours.

The Oromo people are one of the ancient Kushitic speaking indigenous peoples of Africa. According to linguistic studies, archaeological findings and their Oral history, the original home of the Oromo is believed to be the Nubian Valley of the ancient Kushitic empire. They might have organized themselves as a nation, developed their holistic culture, the gadaa, about 4000 B.C. in their original home, then moved southwards and settled in their current regions as a single independent nation. However, the Abyssinian Emperor Minilik II, who got firearms from some European Christian Kingdoms, colonized the Oromo and built the current Ethiopian empire towards the end of 19th century. Since then the Oromo lost their independence, were economically marginalized and culturally segregated. Using this situation as an opportunity, the missionaries of Christianity and Islam expanded their religions and established their own religious institutions in Oromia. On one hand, these two new religions brought new cultural elements that could enrich the Oromo culture. On other hand, they destroyed some valuable Oromo cultural heritages preserved by the Oromo Indigenous Religion (OIR). Their mission agents did not carefully consider cultural elements that could have continued. Instead, they destroyed them indiscriminately. The OIR galmaas (worship houses) were destroyed together with Oromo cultural and symbolic objects, and were replaced by Church buildings and mosques. The Christian missionaries and their local agents assumed the OIR to be idolatry; its feasts, rites, and ritual practices contradictory to the Bible and the Christian faith. Therefore, at the baptismal rite they changed Oromo names to foreign names and discontinued Moggaasaa (OIR child naming rite). Among the Protestant groups of Oromo Christianity (OC), the Oromo’s cultural songs and dances were regarded to be sinful and were therefore abandoned. Oromo cultural foods and drinks were also forbidden. Against such generalized judgment on the OIR this dissertation raises a question: Are OIR elements contradictory or compatible with the OC? To answer this question, I conducted scientific research, compared OIR and OC elements, and produced this dissertation. This work identifies some identical, similar, partially similar and some differing elements in the two Oromo religions (OIR & OC). It is my hope that these findings might contribute to a religious dialogue that aims to create a peaceful religious coexistence in Oromia and beyond.

Hintergrund: Als Teil des Projektes AAA-prevent zur Analyse jugendlichen Alkoholkonsums in Europa widmet sich diese Arbeit der sozialen Ansteckung des Problemtrinkens. Methode: Deskriptiven Analysen der Daten 57.771 12- bis 16-Jähriger aus 25 Ländern folgen Clusteranalysen zur Unterscheidung von Konsummustern und -kulturen. Mehrebenenanalysen (MLA) bestimmen dann den Einfluss sozialer Trinknormen auf individuellen Problemkonsum. Ergebnisse: Fünf Konsummuster auf individueller Ebene werden unterschieden und auf Länderebene zu drei Trinkkulturen aggregiert. Nach der Identifikation von 7.264 (15.1%) Problemkonsumenten, erbringen MLA den signifikanten positiven Zusammenhang zwischen dem Anteil der Problemtrinker eines Landes und dem individuellen Risiko, ein problematischer Konsument zu sein (OR = 1.05). Schlussfolgerungen: Für Jugendliche müssen andere Alkoholtrinkkulturen als für Erwachsene angenommen und der Einfluss sozialer Normen auf individuelles Trinkverhalten beachtet werden.