| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 5.77 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
A segurança de informação tornou-se um aspeto muito importante de qualquer sistema de informação organizacional. As ameaças em sistemas computacionais têm-se revelado cada vez mais inteligentes permitindo contornar as soluções básicas de segurança como firewalls e antivírus. Lidar com vulnerabilidades de software em sistemas computacionais tornou-se um grande desafio para os administradores de sistemas. Com o crescente número de vulnerabilidades descobertas em cada ano, é impossível para os administradores manter os sistemas livres de ações maliciosas.
Os Sistemas de Deteção de Intrusão (SDI) baseados em anomalias, permitem classificar o tráfego monitorizado de uma rede ou operações de um host em atividades normais ou atividades maliciosas. A eficiência da deteção de intrusões irá depender das técnicas utilizadas nestes sistemas.
Este trabalho propõe a aplicação de técnicas machine learning num SDI a desenvolver no âmbito do projeto SASSI (ANI | P2020 17775). O objetivo principal do trabalho consiste na exploração e aplicação de metodologias para deteção e previsão de cyber ataques em sistemas computacionais, de modo a oferecer apoio na tomada de decisão a administradores de sistemas garantindo assim estabilidade e segurança nos sistemas de informação organizacionais.
Information and data security has become a very important part of any organizational information system. Threats in these systems have become increasingly intelligent, so it can deceive the basic security soluions such as firewalls and antivirus. Dealing with software vulnerabilites in computer systems became a major challenge for system administrators. With the increasing number of vulnerabilites discovered each year, it is impossible for administrators to maitain software on their machines, free of malicious actions. Anomaly based Intrusion Detection Systems (IDS), allows monitorized traffic classification or computer systems classification in normal activity or malicious activity. The efficiency of intrusion detection depends on the techniques used in these systems. This work proposes the application of machine learning techniques in IDS to be developed in the scope of SASSI project (ANI | P2020 17775). The main goal of this work is to explore and apply different methodologies to detect and predict cyber attacks in computer systems so it can give support in decision-making to system administrators and guarantee stability and security in organizational information systems.
Information and data security has become a very important part of any organizational information system. Threats in these systems have become increasingly intelligent, so it can deceive the basic security soluions such as firewalls and antivirus. Dealing with software vulnerabilites in computer systems became a major challenge for system administrators. With the increasing number of vulnerabilites discovered each year, it is impossible for administrators to maitain software on their machines, free of malicious actions. Anomaly based Intrusion Detection Systems (IDS), allows monitorized traffic classification or computer systems classification in normal activity or malicious activity. The efficiency of intrusion detection depends on the techniques used in these systems. This work proposes the application of machine learning techniques in IDS to be developed in the scope of SASSI project (ANI | P2020 17775). The main goal of this work is to explore and apply different methodologies to detect and predict cyber attacks in computer systems so it can give support in decision-making to system administrators and guarantee stability and security in organizational information systems.
Description
Keywords
Anomalias Vulnerabilidades Intrusões Segurança Machine learning Classificadores SDI