Refine
Has Fulltext
- yes (103) (remove)
Year of publication
- 2019 (103) (remove)
Document Type
- Doctoral Thesis (103) (remove)
Language
- English (103) (remove)
Is part of the Bibliography
- yes (103)
Keywords
- Klimawandel (4)
- climate change (4)
- Biodiversität (3)
- Spektroskopie (3)
- machine learning (3)
- Anden (2)
- Andes (2)
- Himalaya (2)
- Klima (2)
- Magnetisierungsdynamik (2)
Institute
- Institut für Biochemie und Biologie (20)
- Institut für Geowissenschaften (17)
- Institut für Physik und Astronomie (15)
- Hasso-Plattner-Institut für Digital Engineering GmbH (9)
- Institut für Chemie (9)
- Department Linguistik (8)
- Extern (7)
- Institut für Umweltwissenschaften und Geographie (5)
- Institut für Mathematik (4)
- Sozialwissenschaften (4)
In the era of social networks, internet of things and location-based services, many online services produce a huge amount of data that have valuable objective information, such as geographic coordinates and date time. These characteristics (parameters) in the combination with a textual parameter bring the challenge for the discovery of geospatiotemporal knowledge. This challenge requires efficient methods for clustering and pattern mining in spatial, temporal and textual spaces.
In this thesis, we address the challenge of providing methods and frameworks for geospatiotemporal data analytics. As an initial step, we address the challenges of geospatial data processing: data gathering, normalization, geolocation, and storage. That initial step is the basement to tackle the next challenge -- geospatial clustering challenge. The first step of this challenge is to design the method for online clustering of georeferenced data. This algorithm can be used as a server-side clustering algorithm for online maps that visualize massive georeferenced data. As the second step, we develop the extension of this method that considers, additionally, the temporal aspect of data. For that, we propose the density and intensity-based geospatiotemporal clustering algorithm with fixed distance and time radius.
Each version of the clustering algorithm has its own use case that we show in the thesis.
In the next chapter of the thesis, we look at the spatiotemporal analytics from the perspective of the sequential rule mining challenge. We design and implement the framework that transfers data into textual geospatiotemporal data - data that contain geographic coordinates, time and textual parameters. By this way, we address the challenge of applying pattern/rule mining algorithms in geospatiotemporal space. As the applicable use case study, we propose spatiotemporal crime analytics -- discovery spatiotemporal patterns of crimes in publicly available crime data.
The second part of the thesis, we dedicate to the application part and use case studies. We design and implement the application that uses the proposed clustering algorithms to discover knowledge in data. Jointly with the application, we propose the use case studies for analysis of georeferenced data in terms of situational and public safety awareness.
Sinkholes and depressions are typical landforms of karst regions. They pose a considerable natural hazard to infrastructure, agriculture, economy and human life in affected areas worldwide. The physio-chemical processes of sinkholes and depression formation are manifold, ranging from dissolution and material erosion in the subsurface to mechanical subsidence/failure of the overburden. This thesis addresses the mechanisms leading to the development of sinkholes and depressions by using complementary methods: remote sensing, distinct element modelling and near-surface geophysics.
In the first part, detailed information about the (hydro)-geological background, ground structures, morphologies and spatio-temporal development of sinkholes and depressions at a very active karst area at the Dead Sea are derived from satellite image analysis, photogrammetry and geologic field surveys. There, clusters of an increasing number of sinkholes have been developing since the 1980s within large-scale depressions and are distributed over different kinds of surface materials: clayey mud, sandy-gravel alluvium and lacustrine evaporites (salt). The morphology of sinkholes differs depending in which material they form: Sinkholes in sandy-gravel alluvium and salt are generally deeper and narrower than sinkholes in the interbedded evaporite and mud deposits. From repeated aerial surveys, collapse precursory features like small-scale subsidence, individual holes and cracks are identified in all materials. The analysis sheds light on the ongoing hazardous subsidence process, which is driven by the base-level fall of the Dead Sea and by the dynamic formation of subsurface water channels.
In the second part of this thesis, a novel, 2D distinct element geomechanical modelling approach with the software PFC2D-V5 to simulating individual and multiple cavity growth and sinkhole and large-scale depression development is presented. The approach involves a stepwise material removal technique in void spaces of arbitrarily shaped geometries and is benchmarked by analytical and boundary element method solutions for circular cavities. Simulated compression and tension tests are used to calibrate model parameters with bulk rock properties for the materials of the field site. The simulations show that cavity and sinkhole evolution is controlled by material strength of both overburden and cavity host material, the depth and relative speed of the cavity growth and the developed stress pattern in the subsurface. Major findings are: (1) A progressively deepening differential subrosion with variable growth speed yields a more fragmented stress pattern with stress interaction between the cavities. It favours multiple sinkhole collapses and nesting within large-scale depressions. (2) Low-strength materials do not support large cavities in the material removal zone, and subsidence is mainly characterised by gradual sagging into the material removal zone with synclinal bending. (3) High-strength materials support large cavity formation, leading to sinkhole formation by sudden collapse of the overburden. (4) Large-scale depression formation happens either by coalescence of collapsing holes, block-wise brittle failure, or gradual sagging and lateral widening.
The distinct element based approach is compared to results from remote sensing and geophysics at the field site. The numerical simulation outcomes are generally in good agreement with derived morphometrics, documented surface and subsurface structures as well as seismic velocities. Complementary findings on the subrosion process are provided from electric and seismic measurements in the area.
Based on the novel combination of methods in this thesis, a generic model of karst landform evolution with focus on sinkhole and depression formation is developed. A deepening subrosion system related to preferential flow paths evolves and creates void spaces and subsurface conduits. This subsequently leads to hazardous subsidence, and the formation of sinkholes within large-scale depressions. Finally, a monitoring system for shallow natural hazard phenomena consisting of geodetic and geophysical observations is proposed for similarly affected areas.
A central insight from psychological studies on human eye movements is that eye movement patterns are highly individually characteristic. They can, therefore, be used as a biometric feature, that is, subjects can be identified based on their eye movements. This thesis introduces new machine learning methods to identify subjects based on their eye movements while viewing arbitrary content. The thesis focuses on probabilistic modeling of the problem, which has yielded the best results in the most recent literature. The thesis studies the problem in three phases by proposing a purely probabilistic, probabilistic deep learning, and probabilistic deep metric learning approach. In the first phase, the thesis studies models that rely on psychological concepts about eye movements. Recent literature illustrates that individual-specific distributions of gaze patterns can be used to accurately identify individuals. In these studies, models were based on a simple parametric family of distributions. Such simple parametric models can be robustly estimated from sparse data, but have limited flexibility to capture the differences between individuals. Therefore, this thesis proposes a semiparametric model of gaze patterns that is flexible yet robust for individual identification. These patterns can be understood as domain knowledge derived from psychological literature. Fixations and saccades are examples of simple gaze patterns. The proposed semiparametric densities are drawn under a Gaussian process prior centered at a simple parametric distribution. Thus, the model will stay close to the parametric class of densities if little data is available, but it can also deviate from this class if enough data is available, increasing the flexibility of the model. The proposed method is evaluated on a large-scale dataset, showing significant improvements over the state-of-the-art. Later, the thesis replaces the model based on gaze patterns derived from psychological concepts with a deep neural network that can learn more informative and complex patterns from raw eye movement data. As previous work has shown that the distribution of these patterns across a sequence is informative, a novel statistical aggregation layer called the quantile layer is introduced. It explicitly fits the distribution of deep patterns learned directly from the raw eye movement data. The proposed deep learning approach is end-to-end learnable, such that the deep model learns to extract informative, short local patterns while the quantile layer learns to approximate the distributions of these patterns. Quantile layers are a generic approach that can converge to standard pooling layers or have a more detailed description of the features being pooled, depending on the problem. The proposed model is evaluated in a large-scale study using the eye movements of subjects viewing arbitrary visual input. The model improves upon the standard pooling layers and other statistical aggregation layers proposed in the literature. It also improves upon the state-of-the-art eye movement biometrics by a wide margin. Finally, for the model to identify any subject — not just the set of subjects it is trained on — a metric learning approach is developed. Metric learning learns a distance function over instances. The metric learning model maps the instances into a metric space, where sequences of the same individual are close, and sequences of different individuals are further apart. This thesis introduces a deep metric learning approach with distributional embeddings. The approach represents sequences as a set of continuous distributions in a metric space; to achieve this, a new loss function based on Wasserstein distances is introduced. The proposed method is evaluated on multiple domains besides eye movement biometrics. This approach outperforms the state of the art in deep metric learning in several domains while also outperforming the state of the art in eye movement biometrics.