Modelo predictivo de ciberataques en entornos de internet de las cosas

Data Science allows, in Internet of Things environments, to detect and prevent cyber-attacks using the power of machine learning techniques to autonomously find the best solutions to solve the problems faced by devices in the face of cyber-attacks and vulnerabilities they possess. The CICIOT2023 dataset contains records of the different types of cyber-attacks targeting Internet of Things devices. The objective of the present research is to generate a predictive model by applying machine learning techniques to detect cyber-attacks in Internet of Things environments using the CICIOT2023 dataset, this work highlights the importance of predictive models to protect Internet of Things environments and reduce vulnerabilities by using machine learning and data mining techniques. Classification algorithms, machine learning regression and data mining techniques are applied with training and test data sets to perform predictive modeling of a variety of cyber-attacks. These attacks are categorized into seven families: Distributed Denial of Service (DDoS), Denial-of-Service (DoS), Reconnaissance, Web-based attacks, Brute Force, Spoofing and Mirai. In addition, data visualization algorithms are intended to be used to identify patterns that influence the security of protocols against cyber-attacks. The research results show the importance of a predictive model to maintain the security of IoT devices against the activities of cybercriminals, and thus protect and reduce vulnerabilities in Internet of Things environments. The research work developed can be very useful for those companies specialized in cybersecurity.

• Autor principal: Arias Chevez , German Nelson (author)
• Format: masterThesis
• Idioma: spa
• Publicat: 2024
• Matèries: Internet de las Cosas; Ciberataques; Ciberdelincuentes; Aprendizaje automático
• Accés en línia: https://repositorio.uteq.edu.ec/handle/43000/8129