Browsing by Author "Hoyo, Daniel del"

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • Thumbnail Image
    ItemOpen Access
    Optimization of back-propagation learning algorithm on MLP networks
    (2012) Hoyo, Daniel del
    In order to generate more efficient neural networks, the configuration of the ANN itself has to be optimized, specially refering to its parameters and architecture. To do so, this problem will be approached from the learning and training process point of view, realizing different tests. These evaluations will lead us to determine which are the most optimum parameters for this processes. At the same time, the importance of the input pattern and the data used will be studied, observing how these influences on the learning process, not only from a runtime point of view, but also measuring the obtained error in the trained network. On the other side, the implementation itself will be optimized, doing this by executing the learning algorithm in parallel, using different nodes, meassuring the time needed for completing the trainning, and comparing it with the time needed in a sequential execution.
  • Thumbnail Image
    ItemOpen Access
    Probabilistic map representation using GeoTools
    (2012) Hoyo, Daniel del
    With the current raise of many different smartphones devices and tablets, which have among other functions, a GPS sensor, the ability to make use of services based in the user's location is increasing, as the number of applications offering this type of service. In addition, the number of social networks which allow users to share their location is also increasing. To avoid any kind of leak that could lead to violate user's privacy or security, location privacy techniques should be implemented. Obfuscation will be used in order to avoid the usage of third party solutions. The generation of cloaking regions allows users to define an associated privacy police and configuration, associated with every different type of feature, when map-aware knowledge is available. This paper presents different methods of probabilistic map representations and an obfuscation technique used when unveiling user's position.