[pt] Aumento de eventos climáticos extremos está colocando empresas de seguros em risco, com perdas que chegam a bilhões de dólares. No Sul do Brasil, municípios sofreram perdas devido a eventos climáticos, incluindo um ciclone bomba que causou prejuízos próximos a 2 bilhões de reais. As perdas são em grande parte seguradas, mas avaliar a probabilidade de perdas devido a desastres naturais é difícil devido à dependência intrínseca entre os riscos expostos. Essa dissertação busca estudar ventos extremos na região Sul do Brasil, visando entender como precificar e diluir o risco em áreas de alto impacto. A pesquisa envolve a análise de dados meteorológicos, econômicos, sinistros reportados por seguradoras, prêmios reportados por seguradoras, estrutura populacional, PIB, relevo e outras variáveis relevantes para a pesquisa. O objetivo é estimar cenários de perdas decorrentes de eventos extremos e oferecer informações relevantes para avaliar estratégias de diluição de risco de perdas econômicas. A dissertação mistura distintas áreas, incluindo Economia, Atuária, Ciência de Dados, Estatística e Matemática. / [en] Increasing extreme weather events are putting insurance companies at risk,with losses reaching billions of dollars. In the South of Brazil, municipalities have suffered losses due to climate events, including a bomb cyclone that caused losses of around 2 billion of reais. These losses are largely insured, but evaluating the probability of losses due to natural disasters is difficult due to the intrinsic dependence between exposed risks. This dissertation seeks to study extreme winds in the Southern region of Brazil, aiming to understand how to price and dilute risk in high impact areas. The research involves the analysis of meteorological and economic data, insurance claims reported by insurers, premiums reported by insurers, population structure, GDP, topography, and other relevant variables for the research. The objective is to estimate loss scenarios resulting from extreme events and offer relevant information to evaluate strategies for diluting the risk of economic losses. The dissertation blends distinct areas, including Economics, Actuarial Science, Data Science, Statistics, and Mathematics.

No description available.

Since its launch in October 2014, the Competence Center for Scalable Data Services and Solutions (ScaDS) Dresden/Leipzig carries out collaborative research on Big Data methods and their use in challenging data science applications of different domains, leading to both general, and application-specific solutions and services. In this article, we give an overview about the structure of the competence center, its primary goals and research directions. Furthermore, we outline selected research results on scalable data platforms, distributed graph analytics, data augmentation and integration and visual analytics. We also briefly report on planned activities for the second funding period (2018-2021) of the center.

No description available.

The Plant Hardiness Zone Map consists of thirteen geographical zones that describe whether a plant can survive based on average annual minimal temperatures. As climate change progresses, minimum temperatures in all regions are expected to change. This work programmatically evaluates predicted future climate projection data and converts it to United States Department of Agriculture-defined hardiness zones. Through the next 80 years, hardiness zones are projected to move poleward; in effect, colder zones will lose area and warmer zones will gain area globally. Some implications include changes in crop growing degree days, which could alter crop productivity, migration and settlement of invasive species over native species in shifted zones, and the interruption of plant vernalization, which is an important factor in establishing dormancy. The programmatic evaluation and analysis of hardiness zone change is a strategic lens for viewing the effects and rate of climate change using an easy-to-grasp metric.

The abundance, accessibility, and scale of data have engendered an era where machine learning can quickly and accurately solve complex problems, identify complicated patterns, and uncover intricate trends. One research area where many have applied these techniques is the stock market. Yet, financial domains are influenced by many factors and are notoriously difficult to predict due to their volatile and multivariate behavior. However, the literature indicates that public sentiment data may exhibit significant predictive qualities and improve a model’s ability to predict intricate trends. In this study, momentum SVM classification accuracy was compared between datasets that did and did not contain sentiment analysis-related features. The results indicated that sentiment containing datasets were typically better predictors, with improved model accuracy. However, the results did not reflect the improvements shown by similar research and will require further research to determine the nature of the relationship between sentiment and higher model performance.

No description available.

When attempting to mitigate the spread of an epidemic without the use of a vaccine, many measures may be made to dampen the spread of the disease such as physically distancing and wearing masks. The implementation of an effective test and quarantine strategy on a population has the potential to make a large impact on the spread of the disease as well. Testing and quarantining strategies become difficult when a portion of the population are asymptomatic spreaders of the disease. Additionally, a study has shown that randomly testing a portion of a population for asymptomatic individuals makes a small impact on the spread of a disease. This thesis simulates the transmission of the virus that causes COVID-19, SARSCoV- 2, in contact networks gathered from real world interactions in five different environments. In these simulations, several testing and quarantining strategies are implemented with a varying number of tests per day. These strategies include a random testing strategy and several uniform testing strategies, based on knowledge of the underlying network. By modeling the population interactions as a graph, we are able to extract properties of the graph and test based on those metrics, namely the degree of the network. This thesis found many of the strategies had a similar performance to randomly testing the population, save for testing by degree and testing the cliques of the graph, which was found to consistently outperform other strategies, especially on networks that are more dense. Additionally, we found that any testing and quarantining of a population could significantly reduce the peak number of infections in a community.

Current methods of production forecasting such as decline curve analysis (DCA) or numerical simulation require years of historical production data, and their accuracy is limited by the choice of model parameters. Unconventional resources have proven challenging to apply traditional methods of production forecasting because they lack long production histories and have extremely variable model parameters. This research proposes a data-driven alternative to reservoir simulation and production forecasting techniques. We create a proxy-well model for predicting cumulative oil production by selecting statistically significant well completion parameters and reservoir information as independent predictor variables in regression-based models. Then, principal component analysis (PCA) is applied to extract key features of a well’s time-rate production profile and is used to estimate cumulative oil production. The efficacy of models is examined on field data of over 400 wells in the Eagle Ford Shale in South Texas, supplied from an industry database. The results of this study can be used to help oil and gas companies determine the estimated ultimate recovery (EUR) of a well and in turn inform financial and operational decisions based on available production and well completion data.

My work focuses on taking a single person as input and predicting the intentional movement of one dance partner based on the other dance partner's movement. Human pose estimation has been applied to dance and computer vision, but many existing applications focus on a single individual or multiple individuals performing. Currently there are very few works that focus specifically on dance couples combined with pose prediction. This thesis is applicable to the entertainment and gaming industry by training people to dance with a virtual dance partner. Many existing interactive or virtual dance partners require a motion capture system, multiple cameras or a robot which creates an expensive cost. This thesis does not use a motion capture system and combines OpenPose with swing dance YouTube videos to create a virtual dance partner. By taking in the current dancer's moves as input, the system predicts the dance partner's corresponding moves in the video frames. In order to create a virtual dance partner, datasets that contain information about the skeleton keypoints are necessary to predict a dance partner's pose. There are existing dance datasets for a specific type of dance, but these datasets do not cover swing dance. Furthermore, the dance datasets that do include swing have a limited number of videos. The contribution of this thesis is a large swing dataset that contains three different types of swing dance: East Coast, Lindy Hop and West Coast. I also provide a basic framework to extend the work to create a real-time and interactive dance partner.