Refine
Year of publication
Document Type
- Conference Proceeding (117)
- Doctoral Thesis (22)
- Article (2)
- Book (2)
- Master's Thesis (2)
Language
- English (145) (remove)
Keywords
- Korpus <Linguistik> (29)
- Computerlinguistik (27)
- Computerunterstützte Kommunikation (24)
- Soziale Medien (21)
- Pragmatik (19)
- Linguistik (17)
- Social Media (17)
- NER (13)
- Named entity recognition (13)
- corpus linguistics (11)
Institute
- Informationswissenschaft und Sprachtechnologie (58)
- Informatik (45)
- Übersetzungswissenschaft und Fachkommunikation (17)
- Fachbereich IV (10)
- Fachbereich I (2)
- Fachbereich III (2)
- Musik und Musikwissenschaft (1)
- Psychologie (1)
- Technik (1)
- Universität (1)
This thesis focuses on cosmic rays and Nature of Science (NOS). The first aim of this work is to investigate whether the variegated aspects of cosmic ray research -from its historical development to the science topics addressed herein- can be used for a teaching approach with and about NOS. The efficacy of the NOS based teaching has been highlighted in many studies, aimed at developing innovative and more effective teaching strategies. The fil rouge that we propose unwinds through cosmic ray research, that with its century long history appears to be the perfect topic for a study of and through NOS.
The second aim of the work is to find out what knowledge the pupils and students have regarding the many aspects of NOS. To this end we have designed, executed, and analyzed the outcomes of a sample-based investigation carried out with pupils and students in Palermo (Italy), Tübingen and Hildesheim (Germany), and constructed around an open-ended questionnaire. The main goal is to study whether intrinsic differences between the German and Italian samples can be observed.
The thesis is divided in three parts. In the first part we reconstruct the intricate history of cosmic ray research. First, we present the initial studies that proceeded the discovery of Viktor Hess in 1912, and then the pioneer years of research that unveiled the phenomenological and interpretational features of cosmic radiation. We then continue with the history of the mature phase of cosmic rays research focusing on the discovery and characterization of extensive air showers.
In the second part of the thesis we first present the various aspects of NOS, including the operative definition adopted here and based on the 14 objectives proposed by McComas, Almazroa, and Clough. We then discuss the design, and execution of our sample-based investigation, finally we report in details the results of our analysis, performed with the MAXQDA software program.
In the third part of the thesis the aspects of cosmic ray research, in its historical, technological and cultural developments, are observed through the lens of NOS. We therefore highlight aspects, moments, episodes of cosmic ray research that might elucidate and substantiate, with examples and occasions for discussions, the 14 statements of McComas and collaborators, and we suggest some didactic objectives and units, which can be developed by pupils and teachers
Accessible communication is easy to find, easy to perceive, easy to understand and is related to the previous knowledge of the recipients. Target groups may have special communication needs necessitated by sensory, cognitive or psychological impairment, low education or critical life events such as illness, loss of reference persons or forced migration. When texts fail to address the needs of these target groups, they become barriers to successful interaction: sensory and cognitive barriers, media and culture barriers, language or specialized language barriers or even motoric barriers.
Accessible communication comprises all measures to prevent texts (oral or written) from becoming a barrier for the intended target groups. This is a broad field of action that comprises, but goes way beyond Easy and Plain Language: A text could be written in Easy Language but the intended users have no access to it; a text may not be well perceived and therefore not understood; a text may be accessible online, but directed to users with different media preferences; a text may be easy to perceive and understand but not acceptable with respect to the politeness requirements of the users. What good are such texts to the target audience?
This paper, that was held as Plenary Speech on the 2019 Klaara Network conference, focuses on Easy Language and Accessible Communication in Germany and discusses communication barriers and the conditions of communicative accessibility for people with diverse needs in the light of the following questions: What road have we travelled thus far? Where are we right now? Have we really covered all necessary aspects of accessibility? Where do we go from here?
Many studies reveal that unchangeable static variables, such as prior offending history and membership in high-risk demographic subgroups, consistently predict recidivism. Recent risk assessment research focuses on dynamic attributes — attitudes, values, and interpersonal skills that are modified by new experiences and, thus, may change during the residential stay.
The current work is dedicated to examining one specific dynamic factor – the hostile attributions bias (HAB) – as well as its dynamic change over time. We supposed that the complex nature of cognitive biases can be defined in various ways, e.g. as immanent personality disposition, as social-cognitive interpretation bias, and even as perceptual bias. Therefore, the present dissertational study integrated mixed quantitative and qualitative research, acknowledging that combined approaches are best suitable for assessing complex phenomena in social science research since they can provide real-life contextual understandings and multi-level perspectives on diverse research questions.
We used a multi-method approach to assess HAB as well as multiple statistical approaches to determine which method is most sensitive to changes in the treatment of distorted cognitions. We evaluated the sensitivity to change of three tools (structured questionnaires, semiprojective tool and computer-based perception task) using three statistical methods for identifying aggregate (group ES, Cohen’s d and SRM) and individual changes over time (RCI, individual ES and SEM). The semiprojective method was shown to be sensitive to identifying the largest proportion of change at both aggregate and individual level. At the individual level all three assessment methods showed sensitivity to change. The use of multi-method research is highly relevant for determining intervention changes in corrective settings. Implications for clinical practice, recommendations for future research, and study limitations are discussed.
Multi-method assessment of the hostile attribution bias in juvenile violent offenders: determining the sensitivity to change of three different assessment methods
The study of vegetation-plot data on a broad geographical scale is of increasing importance in vegetation science. It significantly contributes to the transnational characterisation of vegetation types as well as the better understanding of their large-scale patterns and to habitat typologies, which are important for decision-making processes in European nature conservation.
I examined semi-natural, saline and brackish Baltic Sea grasslands which occur on sedimentary flats at the transition between land and sea. Their diverse vegetation is dependent on low intensity grazing (Dijkema 1990). This valuable part of the European cultural landscape (Küster 2004), which is recognized as Annex I priority habitat type (Natura 2000; European Commission 2013), underwent an overall decrease in quality and quantity within the last 150 years, which is frequently related to abandonment. Thus, the coastal grasslands of the Baltic Sea have been assessed as Endangered in the European Red List of Habitats (Janssen et al. 2016).
Within this thesis I (i) developed a proposal to integrate vegetation data using non-standard scales into general vegetation analyses, (ii) characterised the vegetation of Baltic Sea grasslands on transnational level, (iii) regarded them from a North-west European perspective, (iv) discussed their nature conservation aspects on European scale, (v) investigated changes in their plant species composition and discussed its possible relation to cessation of grazing and (vi) formulated a monitoring concept important for management planning in nature conservation.
Machine learning is often confronted with the problem of learning prediction models on a set of observed data points. Given an expressive data set of the problem to solve, using powerful models and learning algorithms is only hindered by setting the right configurations for both. Unfortunately, the magnitude of the performance difference is large, which makes choosing right configurations an additional problem that is only solved by experienced practitioners.
In this thesis, we will address the problem of hyperparameter optimization for machine learning and present ways to solve it. We firstly introduce the problem of supervised machine learning. We then discuss many examples of hyperparameter configurations that can be considered prior to learning the model. Afterwards, we introduce methods on finding the right configurations, especially those methods that work in the scheme of Bayesian optimization, which is a framework for optimizing black-box functions. Black-boxes are functions where for a given input one can only observe an output after running a costly procedure. Usually, in black-box optimization so-called surrogate models are learned to reconstruct the observations to then offer a prediction for unobserved configurations. Fortunately, recent outcomes show that transfering the knowledge across problems, for example by learning surrogates across different data sets being solved by the same model class, shows promising results.
We tackle the problem of hyperparameter optimization in mainly two different ways. At first, we consider the problem of hyperparameter optimization as a recommendation problem, where we want to learn data set features as well as their interaction with the hyperparameter configurations as latent features in a factorization based approach. We build a surrogate model that is inspired by the complexity of neural networks as well as the ability to learn latent embeddings as in factorization machines. Secondly, as the amount of meta knowledge increases every day, surrogate models need to be scalable. We consider Gaussian processes, as they themselves are hyperparameter free and work very well in most hyperparameter optimization cases. Unfortunately, they are not scalable, as a matrix in the size of the number of data points has to be inverted for inference. We show various methods of simplifying a Gaussian process by using an ensemble of Gaussian process experts, which is much faster to learn due to its paralellization properties while still showing very competitive performance.
We conclude the thesis by discussing the aspect of learning across problems in more detail than simply learning across different data sets. By learning hyperparameter performance across different models, we show that also model choice can be handled by the proposed algorithms. Additionally, we show that hyperparameter performance can even be transfered across different problem tasks, for example from classification to regression.
In distributional semantics, the unsupervised learning approach has been widely used for a large number of tasks. On the other hand, supervised learning has less coverage.
In this dissertation, we investigate the supervised learning approach for semantic relatedness tasks in distributional semantics. The investigation considers mainly semantic similarity and semantic classification tasks. Existing and newly-constructed datasets are used as an input for the experiments. The new datasets are constructed from thesauruses like Eurovoc. The Eurovoc thesaurus is a multilingual thesaurus maintained by the Publications Office of the European Union. The meaning of the words in the dataset is represented by using a distributional semantic approach.
The distributional semantic approach collects co-occurrence information from large texts and represents the words in high-dimensional vectors. The English words are represented by using UkWaK corpus while German words are represented by using DeWaC corpus. After representing each word by the high dimensional vector, different supervised machine learning methods are used on the selected tasks. The outputs from the supervised machine learning methods are evaluated by comparing the tasks performance and accuracy with the state of the art unsupervised machine learning methods’ results. In addition, multi-relational matrix factorization is introduced as one supervised learning method in distributional semantics. This dissertation shows the multi-relational matrix factorization method as a good alternative method to integrate different sources of information of words in distributional semantics.
In the dissertation, some new applications are also introduced. One of the applications is an application which analyzes a German company’s website text, and provides information about the company with a concept cloud visualization. The other applications are automatic recognition/disambiguation of the library of congress subject headings and automatic identification of synonym relations in the Dutch Parliament thesaurus applications.
Grazing animals alter natural processes by affecting ecosystems and at the same time fulfilling ecosystem functions, thus they are regarded as ecosystem engineers. Effects of grazing are mainly studied in managed systems, where grazing animals are restricted in their movement and thus limited to certain vegetation types. On the island of Asinara the grazing system is now, due to its history as agro-penitentiary, a natural grazing system with donkeys, horses, goats, mouflons and wild boars. This multitude of grazers poses a challenge for the Asinara National Park and its management. Therefore this dissertation takes an interdisciplinary approach to investigate grazing animals and their interrelations with different components of the island ecosystem to analyse their role on the island and evaluate their effects on the biodiversity. The composition and distribution patterns of the five grazing animal species have been investigated in the context of the land-cover types of the island ecosystem. In addition, the input on the vegetation through endozoochorous seed dispersal by donkeys and goats was analysed, and the impact of grazing animals on dung beetle assemblages was studied in three highly frequented vegetation units, taking into account the intensity of use by the grazing animals. The results derived from this work highlight the importance of studying grazing animals and their interrelations within an island ecosystem. Moreover, the insights given in this thesis concerning the interrelations of grazing animals with different components of the island should open up the view on grazers and their multifaceted effects on the biodiversity, thus leading to management implementations for a sound functioning of the island ecosystem as well as the conservation and maintenance of biodiversity.
According to the cognitive action architecture approach (CAA-A; Schack, 2002, 2004), mental representations play a central role in movement control, as they are hierarchically structured in long-term memory and differentiate due to expertise. By means of structural dimensional analysis - motoric (SDA-M; Schack, 2012), movements are subdivided into basic action concepts (BACs) that are related to each other, to make mental representation structures visible. The method seeks to psychometrically depict the structural dimensions of conceptually ordered motor knowledge, both case-by-case and group-specific, and has been applied to a variety of sports contexts (Land, Volchenkov, Bläsing, & Schack, 2013). Although the examination of mental representations seems promising regarding the gain of insights into the structure of movements, motor learning processes and the relation of cognitive and movement structures, research in the field of educational research has not been conducted so far. Therefore, the aim of this dissertation project is to address the main research question: What influence can the knowledge about mental representation structures of movements have in the context of physical education and the optimization processes of learning and teaching?
The research issue, research program and research questions are presented in the general introduction of this thesis (chapter 1). To answer the main research question, three different perspectives are introduced in the main part.
The first perspective focuses on the structure of motor skills and the methodological approach to acquiring the mental representation structures. The practical implementation of the SDA-M is discussed in a text-image comparison of questionnaire items (chapter 2). The study aimed at determining an appropriate item format for the use of the SDA-M in the context of physical education. In light of the findings, the use of text or combined items is recommended when applying the SDA-M to children or adolescents.
The second perspective focuses on knowledge about the mental representation structure of learners and their execution of the skill. Based on pupils’ mental representation structure, specific instructions are developed that might contribute to the optimization of movement execution (chapter 3). Results revealed that specific instructions based on pupils’ mental representation of a gymnastics skill have a positive impact on their mental representation structure and on their motor performance.
The third perspective focuses on knowledge about the mental representation structure of teachers, here represented by sports students, examining its relation with their evaluation of pupils’ movements (chapter 4). Results indicate that more structured mental representations are linked with a more precise performance evaluation. Therefore, a relation between physical education students’ mental representation structure and their evaluation of gymnastics skills can be assumed.
Conclusively, in the general discussion, the key findings are highlighted, limitations are discussed and an outlook concerning practical implications and further research is provided (chapter 5). In sum, according to the findings of this dissertation, the acquisition and analysis of the structure of mental representations might constitute an effective means of optimizing the quality of physical education and can be seen as important for improving physical education teacher training as well as motor learning processes in general.
Arthropod herbivores act as mediators for effects that cascade up and down the trophic chain. Therefore, herbivory plays an important role for driving ecosystem processes and influencing ecosystem structures and functions. Generally, ecosystem processes are mediated by interactions between organisms. The plant community composition is influenced by competitive interactions among plants, which is affected by herbivore species. Leaf area loss to insects can reduce tree growth, but alters material flows from canopies to forest soils. Therefore, the chemical quality of litter is changed (increases in nitrogen content) through enhanced nutrient cycling rates caused by herbivory.
Climate and microclimate can affect insect physiology and behaviour directly or indirectly through climate-induced changes of host plants. Temperature determines the geographical range, site and timing of activities, success of oviposition and hatching, and the duration of developmental stages of arthropod herbivores. The activity of poikilothermic insects increases with temperature, and therefore growth and consumption rates are enhanced. However, morphological and functional leaf traits that determine host plant palatability often mediate indirect environmental effects on herbivory. Leaf palatability is determined by leaf toughness, nutrients, and defence compounds. In warm environments, expected high rates of arthropod herbivory can then be suppressed by negative changes of leaf traits.
Microclimate gradients are found across the different strata of forest ecosystems. Abiotic factors change vertically between forest layers due to a micro-environmental gradient. Along the vertical gradient, microclimate is affected by the light regime, with increasing temperatures and decreasing humidity from understorey to upper canopies. Various organisms are distributed along the vertical forest gradient based on changes in environmental conditions and in the quality and quantity of available resources. Temperate deciduous forests reveal highly stratified arthropod communities with vertical and horizontal distribution patterns. Microclimatic requirements and the availability of food resources along the vertical forest gradient can reflect spatial distributions and preferences of arthropods.
This research study investigated arthropod herbivory on leaves of deciduous tree species along the vertical gradient of temperate forests. A field study with ten forests sites in Central Germany and an experimental study in greenhouses were conducted, addressing effects of microclimate and leaf traits on arthropod herbivory. Juvenile and adult individuals of Fagus sylvatica L. (European beech), the dominant deciduous tree species in Central Europe, were chosen as main research subjects. Furthermore, Acer pseudoplatanus L. (Sycamore maple) and Carpinus betulus L. (hornbeam), two frequent tree species in the forest understorey, were also surveyed.
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.
