Now showing items 1-15 of 15

  • Adaptive scheduling on power-aware managed data-centers using machine learning 

    Berral García, Josep Lluís; Gavaldà Mestre, Ricard; Torres Viñals, Jordi (IEEE Computer Society Publications, 2011)
    Conference report
    Restricted access - publisher's policy
    Energy-related costs have become one of the major economic factors in IT data-centers, and companies and the research community are currently working on new efficient power-aware resource management strategies, also known ...
  • A defuzzification based new algorithm for the design of Mamdani-type fuzzy controllers 

    Saade, J.J. (Universitat Politècnica de Catalunya. Secció de Matemàtiques i Informàtica, 2000)
    Article
    Open Access
    This paper presents a new learning algorithm for the design of Mamdani-type or fully-linguistic fuzzy controllers based on available input-output data. It relies on the use of a previously introduced parametrized defuzzification ...
  • A multistrategy approach for digital text 

    Castillo, María Dolores del; Serrano Moreno, José Ignacio (Universitat Politècnica de Catalunya. Secció de Matemàtiques i Informàtica, 2005)
    Article
    Open Access
    The goal of the research described here is to develop a multistrategy classifier system that can be used for document categorization. The system automatically discovers classification patterns by applying several empirical ...
  • A post-processing strategy for SVM learning from unbalanced data 

    Núñez Castro, Haydemar; González Abril, Luis; Angulo Bahón, Cecilio (2011)
    Conference lecture
    Open Access
    Standard learning algorithms may perform poorly when learning from unbalanced datasets. Based on the Fisher’s discriminant analysis, a post-processing strategy is introduced to deal datasets with significant imbalance ...
  • Comparison of active learning methods for automatic document classification 

    Marcé Gomis, Marc (Universitat Politècnica de Catalunya, 2018-06-21)
    Master thesis
    Open Access
    Getting correctly labelled data is an important preliminary stage for many supervisedmachine learning problems, but it can also be really difficult to perform. Sometimes, toobtain a good model, it could require tens or ...
  • Detección de transiciones posturales basado en un acelerómetro y un sensor de presión mediante algoritmos de aprendizaje automático 

    Pilataxi Pilataxi, Jhon Ivan (Universitat Politècnica de Catalunya, 2019-01-31)
    Master thesis
    Restricted access - author's decision
    [CASTELLÀ] Los sensores inerciales basados en Sistemas Micro Electro Mecánicos, gracias a su reducido tamaño y bajo consumo de energía, han permitido el desarrollo de sistemas portables orientados al reconocimiento de ...
  • Evolutionaty training for dynamical recurrent neural networks: an application in finantial time series prediction 

    Delgado Calvo-Flores, Miguel; Pegalajar Jiménez, Mª Carmen; Pegalajar Cuéllar, Manuel (Universitat Politècnica de Catalunya. Secció de Matemàtiques i Informàtica, 2006)
    Article
    Open Access
    Theoretical and experimental studies have shown that traditional training algorithms for Dynamical Recurrent Neural Networks may suffer of local optima solutions, due to the error propagation across the recurrence. In the ...
  • Handling binary classification problems with a priority class by using Support Vector Machines 

    Gonzalez Abril, Luis; Angulo Bahón, Cecilio; Núñez Castro, Haydemar; Leal, Yenny (2017-12-01)
    Article
    Restricted access - publisher's policy
    © 2017 Elsevier B.V. A post-processing technique for Support Vector Machine (SVM) algorithms for binary classification problems is introduced in order to obtain adequate accuracy on a priority class (labelled as a positive ...
  • Horn query learning with multiple refinement 

    Sierra Santibáñez, Josefina; Santibáñez Velilla, Josefina (2008)
    Article
    Restricted access - publisher's policy
    In this paper we try to understand the heuristics that underlie the decisions made by the Horn query learning algorithm proposed in [1]. We take advantage of our explicit representation of such heuristics in order to ...
  • Improvement to the cooperative rules methodology by using the ant colony system algorithm 

    Alcalá Fernández, Rafael; Casillas Barranquero, Jorge; Cordón García, Oscar; Herrera Triguero, Francisco (Universitat Politècnica de Catalunya. Secció de Matemàtiques i Informàtica, 2001)
    Article
    Open Access
    The cooperative rules (COR) methodology [2] is based on a combinatorial search of cooperative rules performed over a set of previously generated candidate rule consequents. It obtains accurate models preserving the highest ...
  • Machine learning aplicat a prediccions durant la nit electoral 

    Fernández Bertolín, Sergio (Universitat Politècnica de Catalunya, 2016-10)
    Master thesis (pre-Bologna period)
    Open Access
    The current project explores electoral data from Spanish national elections from 2000 to 2011. Using these, it is developed and tested an algorithm to predict election results with a small error. The analysed prediction ...
  • Máquinas de soporte vectorial sobre conjuntos de datos no balanceados: propuesta de un nuevo sesgo 

    Núñez Castro, Haydemar; González Abril, Luis; Angulo Bahón, Cecilio (2012-09)
    External research report
    Open Access
    En el aprendizaje con conjuntos de datos no balanceados, la máquina de soporte vectorial (SVM) puede exhibir un bajo rendimiento sobre la clase minoritaria ya que, como otras máquinas de aprendizaje, están diseñadas para ...
  • Modificación del sesgo de una SVM entrenada sobre clases no balanceadas 

    Núñez Castro, Haydemar; Angulo Bahón, Cecilio; González Abril, Luis (Universidad de Sevilla, 2011)
    Conference report
    Open Access
    En el área de aprendizaje automático, uno de los problemas que se presenta es el relacionado con las clases no balanceadas. Esto ocurre cuando en el conjunto de datos se dispone de muchos ejemplos de una clase, pero ...
  • Using genetic feature selection for optimizing user profiles 

    Larsen, Henrik Legind; Marín Ruíz, Nicolás; Martín Bautista, Maria José; Vila Miranda, María Amparo (Universitat Politècnica de Catalunya. Secció de Matemàtiques i Informàtica, 2000)
    Article
    Open Access
    Most of the techniques used in text classification are determined by the occurrences of the words (terms) appearing in the documents, combined with the user feedback over the documents retrieved. However, in our model, ...
  • Video Object Segmentation by Tracking Structured Key Points and Contours 

    Caelles Prat, Sergi (Universitat Politècnica de Catalunya, 2016-08)
    Master thesis
    Open Access
    Covenantee:  Eidgenössische Technische Hochschule Zürich
    In this thesis, we tackle the problem of video object segmentation where we have to classify every pixel of every frame in a video sequence into background and foreground classes. Our algorithms fall in the semi-supervised ...