Accepted Papers

 

  • CLASSIFICATION OF FMRI DATA USING DENSITY BASED SUPPORT VECTOR MACHINE
    Zahra Nazari1, Dongshik Kang1, M. Reza Asharif1, Yulwan Sung2 and Seiji Ogawa2,1University of the Ryukyus,Japan and 2Tohoku Fukushi University, Japan
    ABSTRACT
    Interpreting brain images requires analysis of complex and multivariate data. Machine learning algorithms are the most popular and widely used analysis approaches to train classifiers to decode stimuli, behaviours, mental states, and other variables of interest from functional Magnetic Resonance Imaging (fMRI) data and thereby show the data contain enough information about them. In the present study, multivariate statistical pattern recognition methods including Support Vector Machines (SVM) and Density Based Support Vector Machines (DBSVM) were used to classify fMRI volumes. Finally we did a comparative analysis between the results of SVM and DBSVM classifiers.
  • NONLINEAR EXTENSION OF ASYMMETRIC GARCH MODEL WITHIN NEURAL NETWORK FRAMEWORK
    Josip Arneric1 and Tea Poklepovic2,1Department of Statistics,Croatia,2 Department of Quantitative Methods,Croatia
    ABSTRACT
    The importance of volatility for all market participants has led to the development and application of various econometric models. The most popular models in modelling volatility are GARCH type models because they can account excess kurtosis and asymmetric effects of financial time series. Since standard GARCH(1,1) model usually indicate high persistence in the conditional variance, the empirical researches turned to GJR-GARCH model and reveal its superiority in fitting the asymmetric heteroscedasticity in the data. In order to capture both asymmetry and nonlinearity in data, the goal of this paper is to develop a parsimonious NN model as an extension to GJR-GARCH model and to determine if GJR-GARCH-NN outperforms the GJR-GARCH model.
  • OCTONION VALUED NEURAL NETWORK TO FORECAST THE DAILY SOLAR IRRADIATION
    L. Saad Saoud, F. Rahmoune, V. Tourtchine, K. Baddari,University M'hamed Bougara Boumerdes, Algeria
    ABSTRACT
    In this paper, the octonion neural network to forecast the daily solar irradiation is proposed. A method to transform the daily meteorological parameters to octonion numbers is investigated. This method gives the opportunity to forecast the daily solar irradiation using one octonion input rather than six inputs, which decrease the input dimension vector. In addition, it produces eight naturally dimensions to the training algorithm rather than one dimension in the real valued neural networks. The octonion input contains the combination of the complex valued meteorological parameters (the air temperature, the relative humidity, the sunshine duration, the wind speed and speed direction) and time index. Comparison with the real, complex and quaternion valued neural networks for forecasting the daily solar irradiation shows that the proposed method is promising to deal with such problem.

 

Copyright © AIAPP 2016