【摘要】 动态风险识别是根据已知的风险信息尽早预测未来可能存在的风险。基于支持向量机的风险识别技术能够较全面地、自动地通过学习模型来识别可能存在的风险，该技术已经成为动态风险识别的主要方法。为了提高识别的效率与准确性，支持向量机参数的选取非常关键，而人工免疫算法是一种有效的随机全局优化技术，具有精确度高、收敛速度快且不易陷入局部最优解等优点。该文首先对原始数据进行特征选取及降维处理，然后通过人工免疫优化算法(IOA)选择支持向量机(SVM)的惩罚参数和核函数的参数，同时结合支持向量机多分类方法的优势，提出一种新的动态风险识别模型—基于支持向量机和免疫优化算法集成的动态风险识别模型。在Heart-Disease数据集上的实验结果表明，该模型正向与反向的抗原识别率分别为95.82%和96.01%,均高于传统识别模型。更多还原

【Abstract】 Dynamic risk identification is to predict possible future risks as early as possible based on known risk information. The risk identification technology based on support vector machines can more comprehensively and automatically identify possible risks through learning models. This technology has become the main method of dynamic risk identification. In order to improve the efficiency and accuracy of recognition, the selection of support vector machine parameters is quite important, and the artificial immune algorithm is an effective stochastic global optimization technique with high accuracy, fast convergence speed and not easy to fall into local optimal. We firstly perform feature selection and dimensionality reduction processing on the original data, and then use the artificial immune optimization algorithm to select the penalty parameters of the support vector machine and the parameters of the kernel function. Combined with the advantages of the support vector machine multi-classification method, we propose a new dynamic risk identification model based on the integration of support vector machines and immune optimization algorithms. The experiment on Heart-Disease dataset shows that the forward and reverse recognition accuracy rates of this model are 95.82% and 96.01%,respectively, which are higher than that of the traditional SNSA,CSA and DCA models.更多还原