Comparación de técnicas de machine learning para detección de sitios web de phishing

• Autor: Moncada Vargas, Andrés Eduardo
• Formato: tesis de grado
• Fecha de Publicación: 2021
• Institución: Universidad de Lima
• Repositorio: ULIMA-Institucional
• Lenguaje: español
• OAI Identifier: oai:repositorio.ulima.edu.pe:20.500.12724/13842
• Enlace del recurso: https://hdl.handle.net/20.500.12724/13842
• Nivel de acceso: acceso abierto
• Materia: Aprendizaje automático; Suplantación de identidad; Protección de datos; Machine learning; Phishing; Data Protection; https://purl.org/pe-repo/ocde/ford#2.02.04

Phishing is the theft of personal data through fake websites. Victims of this type of theftar e directed to a fake website, where they are asked to enter their data to validate their identity. At that moment, theft is carried out, since entered data are stored and used by the hacker responsible for said attack to sell them or enter to websites and perform a fraud or scam. In order to conduct this work, we researched different methods for detecting phishing websites by using machine learning techniques. Thus, the purpose of this work is to compare machine learning techniques that have demonstrated to be the most effective methods to detect phishing websites. The results show that decision tree classifiers such as Decision Tree and Random Forest have achieved the highest accuracy and efficacy rates, with values between 97% and 99%, in detecting these types of websites.