SciELO - Scientific Electronic Library Online

vol.7 issue3Forecasting of Multiple Points in Time Series with Support Vector Machines author indexsubject indexsearch form
Home Pagealphabetic serial listing  

Services on Demand




Related links

  • Have no similar articlesSimilars in SciELO


Computación y Sistemas

On-line version ISSN 2007-9737Print version ISSN 1405-5546

Comp. y Sist. vol.7 n.3 Ciudad de México Jan./Mar. 2004




The Journal Computación y Sistemas, since its creation, it has been characterized for being a space for the publication of scientific papers, that although is related to the Computing Science, could be part of some other research areas. The original idea to bring an heterogeneous forum around a same area of knowledge human to expose the new scientific results obtained by Iberoamerican researchers, has given results after five years of existence, focus on that central idea, that joint, in multiples perspectives, the excellent results that have been presented in the six volumes published until today. In this number is presented that manifest and cohesion with the papers which are presented by me to the lectors.

The first paper from E. Bautista Thompson, E. Guzmán Ramírez and F. Figueroa Nazuno, presents the evaluation of the prediction of multiple time series, with vectorial support machines, using one technique of the common type of this machines: using a landsline window with multiple functions of Kernel. The support vectorial machines, cradles in the Vapnik Theory, are a kind of learning machines that implement the inductive principle of minimization of structural risk from the statistical learning theory, which actually they have a great application for regression, classification and estimation problems, among others. The authors in this paper show an application of these machines to predict multiple time series.

Cyrus Dana Vesuna and Jean Francois Pâris, from Houston University, in the paper entitled "An Empirical Study of Harmonic Broadcasting Protocols", show the advantages and disadvantages in the use of that kind of protocols for the videos distribution under demand in TCP/IP networks. The harmonic diffusion of protocols are types of protocols of periodic diffusion that are characterized by decrease the channels band width, keeping equal longitudes of the segments, but requiring a relative great quantity of independent data fluids. The authors, in experimental way, show that even those disadvantages, that protocols are a favorable option to be consider for the selection of a video distribution protocol under demand on internet.

The third paper, from Cecil Chow Robilotta, describes the actual techniques fundaments for the obtaining of medical images, particularly in emission tomography for the study of systems and tissues. The author discusses some of the acquisition process and images reconstructions obtained by tomography and present a collection of necessary corrections in these processes.

In "Steinbuch's Lernmatrix: Theoretical Advances", Flavio Sánchez Garfias, Juan Luis Díaz de León and Cornelio Yáñez M., present some results to get a theoretical frame that describes the behavior of Lernmatrix associative memory. The paper is a continue of anothers papers presented by the authors to establish the necessary conditions for the Stainbuch's Lernamtrix recuperate, in perfect form, all its fundamental collection, once that it is get an alternate form to represent that matrix.

J.J. Medel, Pedro Guevara López and Alberto Flores Rueda, in their paper, "Characterization of Real Time Digital Filters to Digital Computers", discuss some important considerations to identify in the implementation of a digital filter in real time, to estimate the parameters of a real system through a motor C.C. model. In the paper, it is proposed a square minimum estimator, to execute with the real time characteristics of digital filter, if are considered the necessary conditions and with the adequate equipment to get that goal.

At last, Juan Martínez Miranda, Arantza Aldea and René Bañares Alcántara discuss a model based in multi agents systems to be utilized in the work team configuration responsabilized with related items. The proposed model includes emotional aspects and personality of the group members. In order to take the proposal ideas in the model, the authors include the description for a first prototype and the obtained results with the simulation.

With the presentation of this papers with scientific quality, the journal Computación y Sistemas demonstrates that keep its original objectives, for which it was created, and invites to the Ibero–American Scientific community and from others countries to continue with their collaborations.


Bárbaro J. Ferro
Associate Editor

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License