Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/96688
Title: Detección de anomalías en entornos del internet de las cosas
Author: Mellizo-Soto Díaz, Gonzalo
Director: Casas-Roma, Jordi  
Tutor: Hernández Gañán, Carlos  
Abstract: In recent years the amount of connected devices has greatly increased, with an increasing number of applications in the industry each day. This devices can be subject of attacks causing instability or data leaks that can be dangerous both for the users and the enterprises, in order to avoid or confront them, security and early detection are becoming a must in a connected world. The focus is the monitoring and detection of the attacks in Internet of Things devices using state of the art Machine Learning techniques. Models such as SVM, DBScan or Isolation Forests have been used and assembled in order to identify with a better accuracy when an attack is happening. With this assembly, attack detection has increased up to 15% comparing to traditional methods and individual model usages and times have been considerably reduced. An active use of Machine Learning models has shown a great improvement at anomaly detection by securing the devices and decreasing the reaction times when facing attacks.
Keywords: machine learning
internet of things
anomaly detection
Document type: info:eu-repo/semantics/masterThesis
Issue Date: 8-Jun-2019
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Bachelor thesis, research projects, etc.

Files in This Item:
File Description SizeFormat 
gmellizoTFM0619memoria.pdfMemoria del TFM13,05 MBAdobe PDFThumbnail
View/Open