SOFM算法将训练样本聚类，然后分别应用SVR来预测股票价格走势。
We use SOFM algorithm to train the samples clustering, and employ SVR respectively to predict the price trend of stock.
依据大量的调查资料，应用有序样本聚类方法，按林分类型、立地条件和林龄来确定次生林改造生长量划分标准；
Based on a large amount of investigating data, the authors use cluster analysis method with ordered samples to set dividing standards for the growth rate of secondary forests rechange.
有效的离散化可以显著地提高系统对样本的聚类能力，增强系统对数据噪音的鲁棒性。
Effective data discretization can obviously improve system ability on clustering instances, and can also make systems more robust to data noise.
鉴于此，本文又提出了一种改进的ART2网络学习算法来实现动态样本的聚类，同时给出了该方法的实验仿真结果。
Whereas, an improved ART2 neural network clustering algorithm is proposed to realize the clustering of dynamic samples, and the simulation results are given out at the same time.
首先，利用“类属函数”对原始样本进行预处理，以提高聚类样本的质量；
At first, the original samples were pretreated by using the membership class function that can improve the quality of cluster sample.
本文提出一种以模式聚类为基础的病态样本判定方法，并给出基于模式相似度计算的投票剔除算法。
The author presented a method for morbid sample recognition that base mode clustering, paper proposed a eliminating algorithm of voting that base mode similarity calculating.
灰色聚类法，不仅能够相当正确地判别结构的可靠性等级，而且在灰色聚类系数矩阵中显示了各样本对于不同鉴定等级的隶属程度。
This method can not only exactly judge the reliability grade of the structure, but also display the subjection of various samples to different grades in grey clustering coefficient matrix.
在模糊c -均值聚类的基础上选择训练样本，可以提高训练样本的准确度，满足了训练样本所需的单一性原则。
Selecting train sample on the basis of fuzzy C-mean clustering can improve accuracy of train sample, singleness of train samples can be satisfied.
该算法将聚类方法和KNN算法的优点结合起来，从而达到缩减了训练样本数量，减少了算法计算量，加快检索速度的目的。
This algorithm combine advantages of KNN and Clustering, decreasing training samples and quantity of algorithm calculating, and increasing the speed of retrieval.
结合均匀设计和聚类思想提出的样本优选方法，在一定程度上解决了神经网络样本选择的问题。
Based on uniform design and the cluster theory, a optimum selecting method of sample is proposed to solve the problem of sample selection.
改进后的聚类结果既消除了采样误差，又保持了云类样本的基本特征属性。
Therefore, the improved FCM clustering results can reduce the sampling errors and retain the main attributes of cloud classification samples.
用平均连锁聚类法构建了样本的遗传相关聚类图。
A cluster dendrogram of the sample was constructed using average linkage clustering.
通过自组织竞争网络的聚类特征，改善样本训练对BP网络性能的影响。
The effect of samples training on BP neural network performance with the clustering characteristic of self-organizing competitive network is improved.
首先利用改进的FCM进行聚类分析，然后将获得的聚类中心作为输入样本，进行KPCA，从而得到主成分图像。
After clustering analysis by the improved FCM, the obtained cluster centers as input samples is used and then the principal component images can be obtained based on KPCA.
本文主要从静态样本和动态样本两方面对动态聚类法进行了研究。
So the paper researches on dynamic clustering method mostly from two aspects — static samples and dynamic samples.
利用最大树法来实现对小样本案例的聚类与提取，避免了制定推理规则的复杂性。
Using maximal tree method to cluster and extract the small cases, it avoids the complexity of establishing reasoning rules.
然后介绍了如何使用模糊聚类算法和等价的前馈神经网络从样本数据中辨识离散的TS模型。
Then we introduce how to identify the TS model from sample data using fuzzy clustering algorithm and equivalent feedforward neural network.
然后根据对多维数据聚类的实验分析结果，通过对样本集的训练进行标识和机器自学习过程来判别异常检测矩阵。
And based on the experimental results of multi-dimensional data clustering, anomaly detection matrix is determined through identifying the training sample set and the machine self-learning.
本文对径向基函数网络提出了一种新的学习算法，利用最小均熵差准则对训练样本进行模式聚类。
This paper presents a new leaning method for radial basis function network, minimum mean entropy difference criterion algorithm is used to get pattern cluster of training sets.
为了解决支持向量机的分类仅应用于较小样本集的问题，提出了一种密度聚类与支持向量机相结合的分类算法。
To solve the problem that support vector machine(SVM) can only classify the small samples set, a new algorithm which applied SVM to density clustering is proposed.
该类学习机利用线性聚类，提取距分类超平面较近的样本构造改进的学习机。
The training data close to the hyperplane are extracted to form the improved learning machines by using linear clustering.
该方法利用模糊似然函数对样本数据进行聚类，并使模糊模型的结构辨识和参数辨识能同时完成，从而实现模糊模型的在线辨识。
The proposed method can accomplish the structure identification and the parameter identification of the fuzzy model in the same time, and implements the on-line identification of the fuzzy model.
引入减法聚类算法对样本数据进行分类，用得到的分类数据对局部模型参数进行离线辨识。
By introducing the subtraction clustering algorithm, the sample data are classified and the local model parameters are identified off-line using the corresponding data set.
通过以全体样本对全体类别加权广义欧氏权距离平方和最小为目标函数，建立了模糊聚类、识别与优选决策统一的理论与循环迭代模型。
With the minimum square sum of weighted Euclidean distances as the objective function, the unified theory and cyclical iteration model of fuzzy cluster, recognition and optimum decision are founded.
标准的FCM算法对大数据样本集进行聚类时极为耗时，而且对噪声比较敏感。
The standard FCM algorithm is not only extremely time-consuming for clustering large data set, but also more sensitive to noise.
然后利用聚类分析方法求得各类样本的聚类中心，得到典型样本。
Then, the center of clustering could be gained by using the method of clustering and the typical sample was obtained.
同时，在模糊C-均值聚类基础上选择训练样本比起直接基于真实地物图选择，减少了主观因素对训练样本选择的影响，因此取得了更高的分类精度。
Selecting train sample on the basis of fuzzy C-mean clustering decreased subjective factor affecting selecting train sample, so higher classification accuracy can be achieved.
本文从不同树种、不同时间、空间的分布出发，应用模糊聚类分区模型先对样本进行分类，以便确定有关参数。
The article applies fuzzy gathering distribution model and distribute species from different trees, time and space decides on the related parameter.
本文从不同树种、不同时间、空间的分布出发，应用模糊聚类分区模型先对样本进行分类，以便确定有关参数。
The article applies fuzzy gathering distribution model and distribute species from different trees, time and space decides on the related parameter.
应用推荐