Adopting machine learning-based IDS systems in SMEs
Jama, Yoonis (2024)
2024
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2024120131723
https://urn.fi/URN:NBN:fi:amk-2024120131723
Tiivistelmä
The rapid advancement of the Internet has transformed the way businesses operate, offering significant advantages but also exposing them to sophisticated cyber threats. Small and Medium Enterprises (SMEs) are particularly vulnerable due to their limited resources and lack of advanced security measures. This thesis investigates the perceived challenges of implementing Machine Learning- based Intrusion Detection Systems (ML-IDS) in SMEs. Utilizing the Technology- Organization-Environment (TOE) framework, this study aims to identify the key barriers and propose practical solutions for enhancing cybersecurity in SMEs.
A mixed-method approach was employed, combining quantitative data from online questionnaires and qualitative insights from semi-structured interviews with cybersecurity and machine learning experts. The findings indicate that the primary challenges faced by SMEs include high costs, complexity of implementation, and compatibility issues with existing IT infrastructure. Additionally, the lack of specialized expertise and the need for continuous updates to keep pace with evolving threats were identified as significant barriers.
The study suggests leveraging cloud services to reduce costs and complexity, outsourcing expertise to mitigate the need for in-house specialists, and adopting a phased implementation approach to ensure smooth integration with existing systems. Continuous training and support for staff are also recommended to maintain the effectiveness of ML-IDS.
By addressing these challenges, SMEs can significantly improve their cybersecurity posture, protecting themselves from sophisticated attacks. This research contributes to the field by providing a comprehensive understanding of the barriers to ML-IDS adoption in SMEs and offering actionable solutions.
A mixed-method approach was employed, combining quantitative data from online questionnaires and qualitative insights from semi-structured interviews with cybersecurity and machine learning experts. The findings indicate that the primary challenges faced by SMEs include high costs, complexity of implementation, and compatibility issues with existing IT infrastructure. Additionally, the lack of specialized expertise and the need for continuous updates to keep pace with evolving threats were identified as significant barriers.
The study suggests leveraging cloud services to reduce costs and complexity, outsourcing expertise to mitigate the need for in-house specialists, and adopting a phased implementation approach to ensure smooth integration with existing systems. Continuous training and support for staff are also recommended to maintain the effectiveness of ML-IDS.
By addressing these challenges, SMEs can significantly improve their cybersecurity posture, protecting themselves from sophisticated attacks. This research contributes to the field by providing a comprehensive understanding of the barriers to ML-IDS adoption in SMEs and offering actionable solutions.