Domain-level Transformation Scenario Security Assessment Based on XGBoost AlgorithmSubtitle as needed

Due to the differences in the security situation of domain-level transformation scenarios fed back by different factors, it is difficult to guarantee the accuracy of security assessment. For this reason, a study on security assessment of domain-level transformation scenarios based on the XGBoost algorithm is proposed. A security situation indicator system covering the entire life cycle was constructed by decomposing the indicator factors that affect the security situation of the scenario, including the generation, modeling, calculation, decision-making, and transmission of transformation scenario information. The XGBoost algorithm was introduced for implementation and evaluation. By constructing a series of decision trees to approximate the objective function - minimizing the parameters consisting of the loss function and the regularization term, the evaluation of the safety state is achieved. In the test results, the R-squared between the evaluation results and the actual situation is good, and the corresponding error is minimal.