Computational Intelligence approaches have demonstrated their ability to make flexible decisions in a dynamic context. Typical examples are fuzzy logic, Recurrent Neural networks, Convolutional Neural Networks, evolutionary computation, artificial systems, and other computer techniques and models. With the aid of these methodologies, decision support systems that provide interlinked solutions for various cyber security and privacy applications could be built that are effective and reliable. Information security and privacy are becoming increasingly crucial with the growth of complicated distributed computing. Large-scale cyberinfrastructure must be protected, so we need adaptable, modifiable, and robust cyber defense systems that can detect a wide range of threats and attacks. These threats and attacks can be external or internal, passive or active, unknown or known etc. Computational intelligence approaches appear promising for improving cyber security systems and are highly being used in the fields of information security and information assurance.
The key areas of advancements in computational intelligence-enabled cybersecurity are discussed in brief. Neural networks: Risk assessment in the Internet of Things (IoT), Anomaly detection, Attack detection, Defense, Detection of data integrity; Evolutionary computation: security problems associated with big data, attack-defense in IoT; Learning theory: extraction of structural features, assurance of data integrity; Fuzzy logic: state-awareness of hybrid IoT systems, trusted communication; Probabilistic methods: detection of activities in cybercriminal networks; e-discovery and digital forensics, cryptocurrency, cybercriminal analysis etc. The advantages of computational intelligence-enabled cybersecurity can be realized in the following three aspects. Cybersecurity solutions that are CI-enabled can meet the urgent need to safeguard data and high-value assets in IoT devices. Intelligently harmful code can be quickly and automatically identified and detected by security systems that support CI. Advanced machine learning algorithms are used by several AI-assisted cyber threat detection systems to find cyber dangers. Insider threats that humans are unable to detect can be identified using CI algorithms. Cybersecurity technologies can now do automated searches, security analyses, and threat identification thanks to the Internet of Things (IoT) and Cloud Computing (CI). By doing this, CI may drastically cut down on human intervention while also quickly learning about potential hazards.
Topics expected but not limited to the following:
- Implementing restricted access in high-security datasets with CI-enabled cybersecurity solutions
- Designing and creating CI-based behavior analysis tools in an IoT environment
- Novel system architectures for integrating CI-enabled cybersecurity in 5G-based solutions
- Methods to overcome security visualization in cybersecurity applications using computational intelligence
- Development of cognitive IoT using CI_enabled cybersecurity systems
- Computer architecture and algorithm security using edge-based AI techniques for cybersecurity
- Cyber investigation implementation for data protection and cyber security monitoring
- Enhanced security risk assessment for cyber-physical-social computing
- Active learning and extraction techniques for CI-based cybersecurity
