For governmental and non-governmental enterprises to tackle risk management with conviction, enterprise management risk assessment (EMRA) is required. This work proposes a research methodology based on an AI-based data mining algorithm (MSVM+EFCNN) for evaluating enterprise-related risks. Initially, all the possible risk assessment indexes of the enterprise are established using a large variety of identification parameters. Then, the data mining algorithms are trained by considering the previous data for building an EMRA model. At last, the current conditions are analyzed using a cluster of risk indicators, and the risk index is identified via the EMRA model. The support vector machine is used for classification purposes, and the fuzzy-based convolutional neural network is enhanced with a genetic algorithm for creating the enterprise risk assessment. The results obtained after keen analysis and experimentation indicate that the data mining algorithms used in this work can evaluate the enterprise-related risks effectively.