Detección de metástasis de cáncer mamario usando máquinas de soporte vectorial a partir de datos de microarray

• Autor: Calderón Niquín, Marks Arturo
• Formato: tesis de grado
• Fecha de Publicación: 2012
• Institución: Universidad Nacional de Trujillo
• Repositorio: UNITRU-Tesis
• Lenguaje: español
• OAI Identifier: oai:dspace.unitru.edu.pe:20.500.14414/4887
• Enlace del recurso: https://hdl.handle.net/20.500.14414/4887
• Nivel de acceso: acceso abierto
• Materia: Microarray, Metástasis de cáncer

Actually the main cause of decease are produce by cancer illness. They can afect a many organs sucha as pancreas, breast, cervix, prostate and others, and breast cancer has high number of cases and this kind of illness has state or level of developing as metastasis, it spreads malign cells for neighbour organs then it produce the highest rate of death. Exists several diagnostic test and one of that is analysis of microarrays is used a different algorithms of machine learning. Support Vector Machine-SVM has obtained a good performance to binary classification and complement of it is a multiple kernel learning-MKL. In this thesis our proposal a new method, it uses a local(datadependent) and nonlinear combination with different kernels. We call it as localized nonlinear multiple kernel learning (LNLMKL). In our experiments for binary microarray classification, different kernels were used in SVM and different kernels combinations. Finally, we report the results of these experiments using eight high-dimensional microarray datasets demostrating that our proposal have performanced better than other methods analyzed.