Web tracking detection using graph neural networks

Web tracking is an extended practice which permeates our day-to-day life, most of the time going unnoticed. In a world where data and information is one of the most valuable assets, the ability to obtain the largest amount of knowledge possible from individuals is desired by a lot of interested enti...

Full description

Bibliographic Details
Author: Gascón Dominguez, Carlos
Format: master thesis
Publication Date:2023
Country:España
Institution:Universitat Politècnica de Catalunya (UPC)
Repository:UPCommons. Portal del coneixement obert de la UPC
Language:English
OAI Identifier:oai:upcommons.upc.edu:2117/401726
Online Access:https://hdl.handle.net/2117/401726
Access Level:Open access
Keyword:Neural networks (Computer science)
detecció de rastrejadors web
xarxes neuronals de grafs
web tracking detection
graph neural networks
Xarxes neuronals (Informàtica)
Àrees temàtiques de la UPC::Informàtica::Arquitectura de computadors
Description
Summary:Web tracking is an extended practice which permeates our day-to-day life, most of the time going unnoticed. In a world where data and information is one of the most valuable assets, the ability to obtain the largest amount of knowledge possible from individuals is desired by a lot of interested entities, ranging from companies to criminals. Tracking techniques pose a threat to our privacy, so research on defensive methods is needed. This thesis explores the creation of a web tracking detector based on the use of graph neural network models. The main idea behind this approach is the fact that the Internet can be naturally expressed as a graph. Therefore, graph neural networks are likely to be a befitting option to exploit the characteristics of the graph representation of the web, enhancing the detection of known and new tracking methods. The main objective of this work is to try to obtain a graph neural network model which is able to correctly classify resources used by web domains into tracking or non-tracking. Other aspects are also studied, in order to have a better understanding of the task.