Rights accessRestricted access - confidentiality agreement
Genetics is said to be the future of medicine, and consequently a lot of research is being focused on it. One of the most promising technologies of this field is DNA microarray, that allows us to measure the expression of thousands of genes in a single experiment of a relatively short time. This process generates huge amounts of raw data that must be treated and analyzed in order to obtain conclusions and knowledge. In this work we adress the feature selection problem applied on this specific field, comparing the performance of three different feature selection techniques that have been classified as embedded techniques: SVM-RFE, l2-AROM and l1-AROM. Moreover we add to this survey a filter technique, known as Golub index. We prove here that simple techniques can present a very good performance, and that techniques that overfitting is one of the main pitfalls of this problem.
All rights reserved. This work is protected by the corresponding intellectual and industrial property rights. Without prejudice to any existing legal exemptions, reproduction, distribution, public communication or transformation of this work are prohibited without permission of the copyright holder. If you wish to make any use of the work not provided for in the law, please contact: email@example.com