Email spam detection using machine learning based text analysis

Phishing is the most common cyber security type of attack that threatens the integrity of all sorts of companies or individuals and is responsible for an incalculable amount of information leakage and economic loss. In this project, an exhaustive investigation has been done regarding the possibility...

Descripción completa

Detalles Bibliográficos
Autor: Nicolas Reyes, Carlos
Tipo de recurso: tesis de maestría
Fecha de publicación:2022
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/386385
Acceso en línea:https://hdl.handle.net/2117/386385
Access Level:acceso abierto
Palabra clave:Computer security
Machine learning
phishing
spam
machine learning
detection
cyber security
detección
Seguretat informàtica
Aprenentatge automàtic
Àrees temàtiques de la UPC::Informàtica::Seguretat informàtica
Descripción
Sumario:Phishing is the most common cyber security type of attack that threatens the integrity of all sorts of companies or individuals and is responsible for an incalculable amount of information leakage and economic loss. In this project, an exhaustive investigation has been done regarding the possibility of using machine learning algorithms to analyze email text, in order to generate models capable of recognizing particular senders or in the contrary, recognize possible impersonation. Moreover, a testing-phased tool has been developed in order to fulfill that purpose, allowing users to authenticate with their gmail account and generate the models that will be used over the desired emails so possible phishing is marked as dangerous in their inboxes.