Cybersecurity Assistant: Neuro-Symbolic Approach to Threat Detection; Cybersecurity Risks with Smart Grids 149 views

Artificial intelligence can now prevent or exploit security vulnerabilities in digital systems. To develop a neuro-symbolic approach to cybersecurity, machine learning and rule-based reasoning were combined to identify and mitigate threats. Models such as random forest and logistic regression were applied to network traffic data to detect suspicious behavior such as repeated login failures or unauthorized IP connections. The product is a desktop application that presents flagged activity in an accessible interface and offers recommended preventive actions. This system demonstrates how integrating symbolic reasoning with machine learning can improve detection accuracy while making cybersecurity tools more approachable for general users. As electric power grids are improved and automated, increasing interconnectivity introduces new risks, particularly through Internet of Things (IoT) devices in the absence of compliance standards. Sustainable and secure smart grids depend on collaboration among policymakers, enterprises, and consumers. Cybersecurity must be integrated into system design rather than treated as an afterthought. In coordination, these projects emphasize that innovation and protection must evolve in parallel to build digital infrastructures that are both intelligent and trustworthy.

BS (Bachelor of Science)

cybersecurity; smart grids; IoT

School of Engineering and Applied Science Bachelor of Science in Computer Science Technical Advisor: Briana Morrison STS Advisor: Peter Norton

English

2025-12-12