This research develops a fuzzy model for risk assessment in information security, suggesting improved accuracy and adaptability in management.
Key Points
The proposed fuzzy risk assessment model achieved classification accuracy of up to 95%, significantly outperforming classical probabilistic models.
Utilizing an adaptive neuro-fuzzy inference system, the model automatically adjusts to dynamic changes in information activity objects.
The approach includes automated generation of rule bases and retraining, enhancing effectiveness in risk management.
Results indicate reduced mean square error to 0.01, highlighting the value of artificial neural networks in minimizing subjectivity and improving assessments.