ID	原文	译文
8954	仿真实验表明,即使在样本中存在目标污染等干扰因素的情况下,所提方法相比传统的利用时域平滑的RBC方法也能获得更好的信噪比改善。	Simulation results show that, even in the samples of target under the condition of interference factors such as pollution, compared with the traditional method using time domain smooth RBC method can get better SNR improvement.
8955	针对指标值为云模型且考虑目标预期和增长预期的多属性决策问题,提出了一种基于前景理论的决策方法。	Expectations for a respect to cloud model and considering the target and growth expectations of multiple attribute decision making problems, this paper proposes a decision-making method based on prospect theory.
8956	提出了增长预期和目标预期的刻画方法,针对以云模型形式给出的属性评价值和双预期值建立了相似性度量方法。	Growth expectations and goals expected characterization method was proposed, according to attribute value are presented in the form of cloud model and double the expected value.
8957	根据累积前景理论和云模型相似度定义了综合前景价值函数,最后根据方案之间的综合前景值总体差异最大化为目标建立权重寻优模型;	A similarity measure method is established based on cumulative prospect theory and defines the prospect of comprehensive cloud model similarity value function, according to the scheme overall comprehensive prospect value differences between maximum weight optimization model is established with the target;
8958	案例说明了方法应用步骤和可行性。	Case shows the feasibility and method application step.
8959	针对支持向量机(support vector machine,SVM)预测过程中影响因素选择、输入特征集优化、核函数选择及参数优化方面存在的问题,提出了一种全过程优化方法。	For support vector machine (support vector machine, SVM) to predict influence factors in the process of selection, the optimal input feature set, choice of kernel function and parameter optimization problems, a process optimization method is proposed.
8960	首先使用频繁模式增长关联规则分析和模糊贝叶斯网络组合模型来解决影响因素选择中存在的主观性问题;	First using frequent-pattern growth association rules analysis and fuzzy bayesian network model to solve the problem of factors influencing the choice of subjectivity;
8961	然后使用在异常值处理和类内距离与类间距离方面进行改进的模糊C均值聚类算法优化输入特征集,减小支持向量机预测模型冗余度及训练样本集过修正度。	And then use the outlier handling and class in the distance and the distance between the class of the improved fuzzy c-means clustering algorithm to optimize the input feature set, reduce the support vector machine (SVM) model for forecasting the redundancy and the degree of training sample set a fixed.
8962	通过比较各核函数的特点选择径向基核函数作为SVC的核函数,改进了粒子群优化算法中微粒速度和位置函数及惯性权重值算法,使用该方法优化SVM参数并建立预测模型。	By comparing the characteristics of the kernel function of radial basis kernel function is chosen as the SVC kernel function, improve the speed and position of particles in particle swarm optimization algorithm function and inertia weight value algorithm, the SVM parameters were optimized using this method and prediction model is established.
8963	最后,通过案例运算和分析,证明该文方法具有更高的预测精度。	Finally, by case calculation and analysis, prove that the method has higher precision of prediction.