2012 | OriginalPaper | Buchkapitel
Parameter Influence in Genetic Algorithm Optimization of Support Vector Machines
verfasst von : Paulo Gaspar, Jaime Carbonell, José Luís Oliveira
Erschienen in: 6th International Conference on Practical Applications of Computational Biology & Bioinformatics
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Support Vector Machines provide a well established and powerful classification method to analyse data and find the minimal-risk separation between different classes. Finding that separation strongly depends on the available feature set. Feature selection and SVM parameters optimization methods improve classification accuracy. This paper studies their joint optimization and attribution improvement. A comparison was made using genetic algorithms to find the best parameters for SVM classification. Results show that using the RBF kernel returns better results on average, though the best optimization for some data sets is highly dependent on the choice of parameters and kernels. We also show that, overall, an average 26% relative improvement with 8% std was obtained.