Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/146208
Title: Análisis de la viabilidad de la aplicación de técnicas de Auto Machine Learning para la detección de intrusiones
Author: Barcina Blanco, Marcos
Tutor: Hernández Jiménez, Enric
Abstract: The purpose of this work is to assess the suitability of applying Auto Machine Learning techniques and tools to the domain of cyber security, specially to network traffic analysis. In order to achieve this, a system that trains several prediction models on a dataset has been developed. The predictions models are trained by using conventional Machine Learning techniques or newer AutoML ones and then the results are compared. The programming language of choice is Python and libraries such as Scikit-learn and TPOT have been imported to implement some classification algorithms. These libraries are open source and their use is very extended for solving Machine Learning problems. The chosen datasets have been previously used in other similar experiments of Machine Learning applied to cyber security. This choice allows for a more in depth comparison between the performance of the current Machine Learning techniques and the actual viability of AutoML.
Keywords: algoritmos de clasificación
machine learning
hyperparameter
AutoML
Document type: info:eu-repo/semantics/masterThesis
Issue Date: 30-May-2022
Publication license: http://creativecommons.org/licenses/by-nc-sa/3.0/es/  
Appears in Collections:Trabajos finales de carrera, trabajos de investigación, etc.

Files in This Item:
File Description SizeFormat 
mbarcinabTFM0622memoria.pdfMemoria del TFM1,78 MBAdobe PDFThumbnail
View/Open