Development of methods and algorithms for intrusion detection and prevention systems based on statistical methods and sustainable machine learning algorithms

Aug 5, 2026

Upload by Asher

Research on vulnerabilities in intrusion detection systems (IDS/IPS) is addressed through statistical methods combined with machine learning to improve detection accuracy as cyber threats continue to grow. The work evaluates IDS/IPS using real cyber-attack data, analyzes common attack types and weaknesses, and develops a test environment to study algorithm effectiveness. Practical value includes stronger protection of information systems, improved defense against advanced hidden attacks, and training support for information security specialists.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages116

Reading Time4 h 52 min

AccessFree to download

Introduction

Intrusion detection systems (IDS/IPS) that utilize algorithms based on statistical methods and machine learning

Comparative analysis of existing machine learning methods for assessing IDS/IPS vulnerabilities

Ministry of Science and Higher Education of the Russian Federation Federal State Autonomous Educational Institution of Higher Education «Ural Federal University named after the first President of Russia B.N. Yeltsin» Engineering School of Information Technologies, Telecommunications and Control Systems School of Professional and Academic Education ADMIT TO THESIS DEFENSE BEFORE THE SEC Head of EP 09.04.03 M.A. Medvedeva «01» June 2024 MASTER THESIS Development of methods and algorithms for intrusion detection and prevention systems based on statistical methods and sustainable machine learning algorithms Research supervisor: Medvedev M.A. Associate Professor, signature Candidate of Economic Sciences Research supervisor: Agbozo E. Senior lecturer, signature Student: Kirin E.D. Group number RIM-210980 signature Yekaterinburg ABSTRACT Topic of master’s thesis: Development of methods and algorithms for intrusion detection and prevention systems based on statistical methods and sustainable machine learning algorithms The master thesis has been written on 114 pages and contains 14 tables, 56 figures, 42 references. Researching vulnerabilities in intrusion detection systems (IDS/IPS) using algorithms based on statistical methods and machine learning is a pertinent topic due to the continuous rise of cyber threats, the necessity of data privacy protection, the application of cutting-edge technologies, and the widespread use of machine learning methods in the field of information security. The practical significance of this research lies in the following aspects: the findings will help identify vulnerabilities in intrusion detection systems (IDS/IPS), thus enhancing the overall security level of information systems; studying algorithms based on statistical methods and machine learning will facilitate the development of new methods for defense against attacks and their integration into existing IDS/IPS systems; the obtained results can be utilized for training information security specialists, thereby enhancing qualification levels and preparing personnel in this field. The economic efficiency of the research directions can be assessed as follows: the use of improved protection algorithms will mitigate cyber attack risks, data breaches, and other incidents, consequently reducing organizational losses associated with security breaches; ensuring reliable protection of information systems from external threats enhances organizational reputation, increases customer and partner trust, potentially leading to expanded business activities and attracting new clients. The scientific novelty proposed by this research involves refining existing vulnerability detection algorithms in IDS/IPS systems, based on a combination of statistical methods and machine learning techniques. Additionally, the feasibility of applying machine learning methods to detect hidden and advanced attacks, which traditional IDS/IPS systems may overlook, will be explored. The research will utilize a wide range of data, including real cyber attack data for experimental testing. Previous research findings in cybersecurity and intrusion detection systems will also be examined. The object of the research is intrusion detection systems (IDS/IPS) utilizing algorithms based on statistical methods and machine learning techniques. The focus of the study is on the system itself, its components, detection algorithms, and mechanisms, as well as its operational principles in the context of identifying vulnerabilities and potential attacks. The subject of the research includes intrusion detection algorithms in IDS/IPS based on statistical methods and machine learning. Based on existing research and literary sources, the research objectives and goals have been identified. The main goal is to evaluate IDS/IPS based on statistical methods and machine learning. The project tasks include: 1. Studying intrusion detection algo

FAQs

What is the main goal of the master thesis?

To evaluate IDS/IPS based on statistical methods and sustainable machine learning algorithms.

How does the thesis assess vulnerabilities and effectiveness of detection algorithms?

It develops a test environment, studies intrusion detection algorithms, and evaluates their effectiveness using a wide range of data including real cyber attack data.

What practical significance does the research provide?

It helps identify vulnerabilities in IDS/IPS to enhance information security, supports the development and integration of defense methods against attacks, and can be used for training information security specialists.

Aug 5, 2026

Upload by Asher