并通过最小化分类误差准则最大化SVM两类输出值概率分布间的距离。
The proposed scheme also employs the rule that minimizes the error of classifications to maximize the distance of the output distributions of two classes.
这是因为考虑到类别的关联性，不同的类别对最小化分类误差的贡献是不同的。
The reason is the contributions of different labels to minimizing the classification error are different due to the inherent label correlations.
讨论了离散贝叶斯分类算法之后，推导了离散贝叶斯分类器的分类误差估算公式。
The algorithm of discrete Bayes classifier is proposed. Then, formulas for estimating classifying error of Bayes classifier are deduced.
支持向量机的分类误差与概率神经网络相近，但支持向量机的使用较概率神经网络简单。
The generalization error of Support Vector Machine is approximately equal to that of Probabilistic Neural Network. And Support Vector Machine is easier to use than Neural Networks.
属性误差主要来源于拓扑分析误差、数据分类误差、文本文件数据的误差和人工解译误差。
The property errors are mainly from errors of topology analysis, data classification, document data and artificial translation.
由确定性退火技术构造学习规则用于优化分类器参数，目的是减少分类误差以及待识别空间的系统熵。
Learning rules are constructed according to deterministic annealing to optimize classifier parameters, on purpose to reduce classification error and system entropy of the space to be identified.
摘要：综合考虑神经网络分类误差率以及训练速率，文中从组合分类器结构出发，提出一种树形多层的BP—LV Q神经网络组合分类器模型。
Absrtact: By considering the error rates and the training speed of neural networks, a hierarchical classifiers which is called as BP - LVQ neural network combination model is proposed in this paper.
有一些实际问题是允许一定程度的分类误差的，为了解决这些问题，本文将基于覆盖的粗糙集模型作了相应的推广，提出了基于覆盖的程度粗糙集模型。
Some practical problems allow certain degree error of classification, in order to solve the problems, this paper extended our model and put forward the degree rough set model based covering.
该方法与人工分类、神经网络、决策树等方法比较，其测试误差低，测试速度高。
Compared with artificial classification, nerve network, and decision tree, its test error is low and the speed is high.
但是，由于不同极化状态回波信号之间的相关性，极化数据不可避免地产生数据冗余，反而增大了岩性分类的误差。
But meanwhile, the data fusion coming from the coherence of different polarizations in the polarimetric data produces more error for lithological classification.
文章在深入分析基本BP算法的基础上，采用四层BP网络，利用误差逆传播学习方法实现了对模式的分类。
This paper classifies patterns by using four-layer BP network and the learning method of error inverse propagation.
简要介绍了材料热膨胀系数的定义及分类，并对经常使用的零件热变形的计算方法及计算产生的误差进行了分析。
The definition and classification of material's thermal expansion coefficient are introduced briefly. The thermal deformation calculation method and error of parts often used in practice are analyzed.
首先，我们说明了H。263句法元素的误差敏感性并根据其敏感性将其进行分类。
Firstly, we investigated the error sensitivities of H. 263 syntactical elements and grouped them into several classes of different significance.
阐述按不希望有的信号所产生误差的分类法。
The classification of the errors due to unwanted signals is described.
本文对测量误差和测量不确定度的定义、来源及分类进行描述。
This article defined the two different concepts: error and uncertainty of measurement, explained how they preduce and listed their types.
首先，本文简单介绍了电压互感器的原理，描述了电压互感器的分类、基本参数和误差分析。
At first, in this paper, the principle of voltage transformer is introduced, and the classification, basic parameters and error analysis of voltage transformer are stated.
此外，提出用带输出误差死区的混合BP算法训练神经元分类器，提高了网络学习训练速度和分类准确性。
Further, a hybrid BP algorithm with dead interval of error is derived for training the neural classifier in order to increase training speed and classification accuracy.
针对这些问题，基于最小化学习误差的增量思想，该文将学习型矢量量化（LVQ）和生长型神经气（GNG）结合起来提出一种新的增量学习型矢量量化方法，并将其应用到文本分类中。
To solve these problems, based on minimizing the increment of learning errors and combining LVQ and GNG, the authors propose a new growing LVQ method and apply it to text classification.
介绍了并联机床误差的基本分类，主要分为静态误差和动态误差两大类。
The basic error classification of the PMT is introduced, including static error and dynamic error mostly.
此分类算法首先计算未知类别样本的重构系数，定义一种误差作为判别标准，根据此误差的大小判断样本的类别归属。
This algorithm firstly computes the reconstruction weights of unknown samples. Then an error, on which the class of samples can be decided based, is defined as a criterion.
只要预处理、特征抽取及分类得当，时间间隔和实验误差等因素对运动想象识别的影响不大。
With the suitable pretreatment, features extraction and classifier design, the influence, which arose from time interval, experiment error and so on, can be nearly ignored.
样本输出结果较理想值误差较小，分类器的识别结果完全符合实际情况。
The errors between output and ideal results of training samples were smaller.
该方法采用纠错编码支持向量机的多类分类技术，降低了经验风险，能对误差进行自动修正，有效地提高了识别率和识别速度。
A kind of error correction coding methods in the communication between singlechip′s serialport and PC in the electric spark forming machine control system is proposed.
再通过分类K L变换去除量化误差图像的谱间相关性，对K L变换后的特征图像采用预测树的方法进一步去除谱间结构相关性和空间相关性。
Then the classified K_L transform is used to reduce the spectral correlation of quantization error image. Finally, the prediction tree is adopted to reduc…
就李睿同志分析分类检索误差的错误及蒋鸿标同志提出“一卡多置不可取”的错误作了分析和阐述。
In this article, the author points out Li Rui mistakes in discussing indexed searching and questions Jiang Hongbiao argument that necardmul tiplacements is undesirable.
介绍了并联机床误差的基本分类和影响并联机床加工精度的误差来源。
The basic classification of parallel machine tool error and error sources which influence the precision of parallel machine tool were introduced.
介绍了并联机床误差的基本分类和影响并联机床加工精度的误差来源。
The basic classification of parallel machine tool error and error sources which influence the precision of parallel machine tool were introduced.
应用推荐