Enviaments recents

  • Question classification in the medical domain using convolutional neural networks and word embeddings 

    Yamaui Gonzalez, Alexandra Carolina (Universitat Politécnica de Catalunya, 2018-10-15)
    Projecte Final de Màster Oficial
    Accés obert
    Question Answering Systems have become important due to the necessity to retrieveanswers from questions stated in Natural Language, without the need of queryingstructured data sources. One of the key steps of these systems ...
  • Analysis and simulation of data prefetching algorithms for last-level cache memory 

    Escuín Blasco, Carlos (Universitat Politècnica de Catalunya, 2018-06-25)
    Projecte Final de Màster Oficial
    Accés obert
    Realitzat a/amb:  České vysoké učení technické v Praze
    Analysis and simulation of data prefetching algorithms for last-level cache memory. Analysis and comparison of one of the latest data prefetching algorithms in terms of performance, network utilization and prefetching accuracy.
  • Hardware Error Injection, Analysis and Tolerance at Operating System Level 

    Bagdasar, Sergiy (Universitat Politècnica de Catalunya, 2018-06-25)
    Projecte Final de Màster Oficial
    Accés obert
    We present a software approach to hardware error injection in a running process on Linux. Then we analyze the existing hardware error reporting tools on Linux and the error tolerance techniques. We propose a new approach ...
  • Improving the scheduler of the OmpSs-2 runtime 

    Marí Barceló, Marc (Universitat Politècnica de Catalunya, 2018-06-27)
    Projecte Final de Màster Oficial
    Accés obert
    In a task-based model such as OmpSs-2, task creation and scheduling are critical to achieve good performance. For this reason, these areas have been redesigned in the Nanos6 runtime to make them scalable and efficient for ...
  • Reducing waiting times and crowding in hospital emergency departments using Machine Learning 

    Casola, Silvia (Universitat Politècnica de Catalunya, 2018-06-27)
    Projecte Final de Màster Oficial
    Accés obert
    Waiting times are linked to risks for the patient and higher mortality rate. We propose a model to reduce waiting times in the ER by using the patient's previous history data from various heath services.
  • Semantic IoT for reasoning and BigData analytics 

    Van Hille, Reinout (Universitat Politècnica de Catalunya, 2018-06-28)
    Projecte Final de Màster Oficial
    Accés obert
    Recent developments in the IoT industries have led to an increase in data availability that is starting to weight heavily on the traditional idea of pushing data to the Cloud. This study focuses on identifying tasks that ...
  • Defining corporate credit ratings for Spanish companies using machine learning techniques 

    Busquet Seguí, Francesc (Universitat Politècnica de Catalunya, 2018-06-11)
    Projecte Final de Màster Oficial
    Accés obert
    The financial Crisis of 2008 exposed the dubious practices followed by the major Credit Rating agencies, which because of conflicts of interests set the highest ratings available to products that presented major risks, ...
  • On explainability of deep neural networks 

    Parafita Martínez, Álvaro (Universitat Politècnica de Catalunya, 2018-06-15)
    Projecte Final de Màster Oficial
    Accés obert
    Deep Learning has attained state-of-the-art performance in the recent years, but it is still hard to determine the reasoning behind each prediction. This project will cover the latest advances on interpretability and propose ...
  • Methods for removing non-interesting itemsets when mining electronic healthcare records 

    Genna, Vincenzo (Universitat Politècnica de Catalunya, 2018-06-27)
    Projecte Final de Màster Oficial
    Accés obert
    Extracting not trivial informations from healthcare data can help doctors to better choose the direction of the research or of a cure. We propose a method to filter not interesting associations among diagnostics and a ...
  • Model-Agnostic process modelling 

    Sànchez Ferreres, Josep (Universitat Politècnica de Catalunya, 2018-06-27)
    Projecte Final de Màster Oficial
    Accés obert
    Modeling techniques in Business Process Management often suffer from low adoption due to the variety of profiles found in organizations. This project aims to provide a novel alternative to BPM documentation, ATD, based on ...
  • Analysis of kernel matrices and their relation to SVM performance 

    Wiejacha, Malgorzata (Universitat Politècnica de Catalunya, 2018-06-18)
    Projecte Final de Màster Oficial
    Accés obert
    The thesis explores the relationship between kernel matrix properties and RVM model performance in terms of its generalization power. It pinpoints desirable qualities of a kernel matrix and provides a heuristic for kernel ...
  • Enhancing the interoperability between distributed-memory and task-based programming models 

    Sala Penadés, Kevin (Universitat Politècnica de Catalunya, 2018-06-27)
    Projecte Final de Màster Oficial
    Accés obert
    Realitzat a/amb:  Barcelona Supercomputing Centre (BSC)
    Hybrid applications allow to exploit both inter- and intra-node parallelism, however the programming models currently used are not designed to be combined. For this reason, we propose a generic mechanism to enhance the ...

Mostra'n més