Abstract: In this novel research a hybrid algorithm is presented, one capable of selecting a set of features that makes it possible to identify individuals who are healthy and those who suffer from prostate cancer. In this research the selection of features is carried out by means of evolutionary algorithms. In previous works, algorithms of this nature have proven their ability to find solutions for optimization problems in a wide range of fields. The present research proposes a novel hybrid algorithm based on genetic algorithms and support vector machines have been developed in order to find the best variables subset for classifying individuals. The results obtained show how well the method performs in comparison to other methodologies. Cases and controls belong to the study MCC-Spain.

Autoría

Descargar referencia