In the fast-paced world of maritime operations, where data flows as freely as the tides, ensuring cybersecurity is paramount. A recent study by Hüseyin Eski from Sakarya University’s Faculty of Technology, Department of Computer Engineering, sheds light on how artificial neural networks (ANNs) can be harnessed to predict and manage network traffic more efficiently. Published in the Sakarya University Journal of Computer and Information Sciences, the research offers a glimpse into the future of maritime cybersecurity.
Eski’s work focuses on the Turkey Maritime Enterprises Inc. (TME), where a security firewall monitors all internet traffic, categorizing it as either “allowed” or “blocked.” The study collected log records from the firewall over five days, extracting 26 variables to identify the most significant parameters impacting the output. Using linear regression (LR) and principal component analysis (PCA), Eski pinpointed the key variables that influence the firewall’s decisions.
The research then took a step further by using these variables to predict the firewall’s output using ANNs. The results were promising, with prediction accuracy ranging between 85% and 88%, and precision between 92% and 98%. The F1-score, which balances precision and recall, also showed impressive results between 89% and 92%.
“ML and PCA methods successfully identified crucial variables for estimating output and reduced the number of variables from 26 to 5,” Eski noted. This reduction in feature space not only improved the ANN’s performance but also made the inputs more interpretable, supporting accurate firewall-traffic prediction.
For the maritime sector, the implications are significant. As ships become more connected, with systems relying on internet-based communication and control, the volume of data traffic is set to rise exponentially. Efficiently managing this traffic is crucial for maintaining operational efficiency and security.
The study’s findings offer a roadmap for maritime companies to enhance their cybersecurity measures. By identifying key variables that influence network traffic, companies can optimize their firewalls to handle increasing data flows more effectively. This can lead to improved operational efficiency, reduced downtime, and enhanced security.
Moreover, the use of ANNs in predicting network traffic opens up new opportunities for proactive cybersecurity measures. By anticipating potential threats, maritime companies can take preemptive actions to mitigate risks, ensuring the safety and security of their operations.
In the words of Eski, “Reducing the feature space from 26 to 6 variables identified via MLR+PCA improved ANN performance across hours.” This underscores the potential of advanced data analysis techniques in revolutionizing maritime cybersecurity.
As the maritime industry continues to embrace digital transformation, research like Eski’s will play a pivotal role in shaping the future of cybersecurity. By leveraging the power of ANNs and advanced data analysis, maritime companies can navigate the digital seas with confidence, ensuring safe and efficient operations in an increasingly connected world.