最佳集合样本数 the optimal number of ensemble samples
If there are noise or contradictory information, results prediction based on small sample set will be greatly influenced.
若在小样本集合中存在噪音或矛盾信息,则对小样本预测的结果会产生很大地影响。
参考来源 - 基于粗糙集理论与支持向量机的数据挖掘方法算法研究·2,447,543篇论文数据,部分数据来源于NoteExpress
以所选模型参数代表点代入地震动随机函数模型,即可以得到地震动时程样本集合。
Substituting the represented points in the proposed random function model and performing the inverse Fourier transform yield the assembles of the stochastic accelerograms.
这种算法非常简单,且在一定条件下,能将样本集合中的所有输入、输出数据对拟合至任意精度。
The algorithm is quite simple, especially, on the certain condition, it can match all the input-output pairs in the training set to any given accuracy.
结果表明该分类器在标准文本样本集合上的性能好于其他三种分类器,在小样本分类上具有一定优势。
The results show that it has better performance than the other three classifier on the standard text sample set, and it has some superiority on small set of samples.
应用推荐