Аңдатпа
The electronic document management system (EDMS) is among the most security-critical components of a university's information infrastructure, because it stores legally significant records and personal data. Conventional risk assessment of such systems relies on checklist audits whose qualitative verdicts cannot adequately capture the ambiguity of cyber threats. This study proposes a fuzzy expert system, ISRISKS, that evaluates the cybersecurity risk of an EDMS by encoding seventy-nine information-security risk factors into sixteen interconnected Mamdani fuzzy machines. Linguistic variables – low, moderate and high – are modelled with triangular membership functions and aggregated through centroid defuzzification into a single security score in the interval from zero to one. The framework was applied to six university platforms, with the Directum document management system as the focal case, and benchmarked against the NIST 800-30, ISO/IEC 27001 and BS 7799 standards together with an independent expert assessment. The Directum EDMS obtained a fuzzy score of 0.954, the highest among the evaluated platforms, confirming a mature security posture. Pearson correlation between the proposed model and the established standards ranged from 0.994 to 0.996, and a one-sample t-test (d.f. = 3, t-crit = 3.18) returned p-values between 0.66 and 0.98, so the null hypothesis of equal means could not be rejected. The results show that the fuzzy expert system matches assessments made to the standard, and that it does so while dealing openly with the uncertainty that EDMS security evaluation always carries.