并提出改进的PRN组合特征选择算法。
所以，对特征选择算法的研究是十分必要的。
信息增益是最有效的特征选择算法之一，但计算量大。
Information gain is one of the most effective feature selections, but its computation is too huge.
针对不平衡数据集，提出一种基于后验概率的特征选择算法。
In this paper, a posterior-probability-based feature selection algorithm is proposed for imbalanced datasets.
本文介绍两种线性特征选择算法，并用于北京地区强震危险区划的研究。
This paper presents two algorithms of linear feature selection used for the study of strong earthquake zoning in Beijing area.
实验结果表明，该算法有效改进迹比准则特征选择算法，同时降低错分率。
Experimental results show that the proposed algorithm improves trace ratio criterion significantly and lower error rate can be achieved at the same time.
本文提出了一种基于相似度的特征选择算法和适应模糊学习算法来实现分类。
In this paper, a web page classification with feature selection and fuzzy learning is proposed.
所提出的特征选择算法不仅提高了车牌字符识别率，而且识别结果十分稳定。
What's more the experimental results indicate that the vehicle character recognition rates are improved significantly and the classification results are very robust using the proposed method.
通过实验准确率选择适用于上市公司财务指标的离散化算法、特征选择算法。
Through experiments choose listed companies financial indicatorsapplicable to the algorithm, experiment with different rating methods and certificateclassification tree superiority.
利用最大余量原理的特征选择算法在目前的机器学习研究中已经占据了重要的地位。
Margin-based algorithms which use the large margin principle for feature selection have already played a crucial role in current machine learning research.
尽管出现了大量的特征选择算法，特征选择仍然面临着新的挑战：如何处理高维海量的样本。
Although there exist numerous feature selection algorithms, new challenging research issues arise for feature selection: from handling a large dimensionality huge number of samples.
作者从基于分辨矩阵的粗糙集属性约简中受到启发，提出了一系列基于粗集理论的文本特征选择算法，即DB1、DB2、LDB。
The author gains insights from attribute reduction based on discernability matrix and proposes a few rough-set based text feature selection algorithms, i. e. , DB1, DB2 and LDB.
兼顾可靠性和精度两方面的要求，提出了一种匹配特征区域的快速最优选择算法；
Based on the needs of reliability and accuracy, a fast optimal matching area selection method of matching feature was presented.
介绍了手写签名特征抽取和选择的一般方法，并在此基础上，提出了一种实用的签名笔迹特殊点提取算法。
In this paper, the method for feature extraction and selection of handwriting signature is discussed. On this basis, a functional algorithm of special points extraction has been presented.
为了提高人工免疫系统中抗体生成速度，基于免疫系统的混沌特征，提出了混沌否定选择算法。
To increase antibody generating speed in artificial immune system, a chaos negative selection algorithm was presented which is based on the chaotic character of immune system.
实验结果表明所采用的算法用于特征选择具有搜索能力强、分类精度高的特点，完全可以替代传统的SFFS算法。
Experimental results prove the ralidity of this algorithms, which can be regarded as a valid alternative to classical SFFS method.
针对基于基因表达数据的分类，本文从特征基因选择和支持向量机分类算法两个方面进行了改进。
This thesis improves classification using gene expression data method in two aspects: feature selection and SVMs classification algorithm.
在特征向量的选择中，本算法还用到了训练集的类别标签和类别平均向量的判别信息。
In selecting vectors, the algorithm also USES the class label of training set and the judgment information of class mean vector.
文本分类的两个重要的研究方向是：特征选择与文本分类算法。
Two important research directions of text classification are: feature selection method and text classification algorithm.
在特征选择阶段，采用计算近似增益的平行算法，解决模型计算量过大和系统开销问题。
In the feature selection stage, the computing approximate gains in parallel is adopted in order to solve the computational expensiveness of the model and system spending.
传统的K-均值算法选择的相似性度量通常是欧几里德距离的倒数，这种距离通常涉及所有的特征。
The Euclidean distance is usually chosen as the similarity measure in the conventional K-means clustering algorithm, which usually relates to all attributes.
传统的K-近邻算法选择的相似性度量通常是欧几里德距离的倒数，这种距离通常涉及所有的特征。
The Euclidean distance is usually chosen as the similarity measure in the conventional K-NN algorithm, which usually relates to all attributes.
算法所用特征对平坦衰落信道具有衰落不变性，对频率选择性衰落信道也具有很强的抗多径能力。
The feature used in the algorithm is invariant with respect to flat fading, and can resist multipath fading of frequency-selecting channel in quite a few degree.
而后选择了一种在特征空间进行非监督学习的算法，对彩色图像进行了分割，实验证明该方法识别率很高。
Then it studies segment color images in the feature space to choose a method which bases on non-supervise, and the experiments have testified the efficiency.
算法所用的小波变换相位信息特征对平坦衰落信道具有衰落不变性，对频率选择性衰落信道也具有很强的抗多径能力。
The phase information of wavelet transform used in the algorithm is invariant with respect to flat fading, and can resist multipath fading of frequency-selecting channel in quite a few degree.
算法所用的小波变换相位信息特征对平坦衰落信道具有衰落不变性，对频率选择性衰落信道也具有很强的抗多径能力。
The phase information of wavelet transform used in the algorithm is invariant with respect to flat fading, and can resist multipath fading of frequency-selecting channel in quite a few degree.
应用推荐