位置：成果数据库 > 期刊 > 期刊详情页
一种基于稀疏表示系数的特征相关性测度
  • 期刊名称：模式识别与人工智能
  • 时间：0
  • 页码：-
  • 分类：TP391[自动化与计算机技术—计算机应用技术;自动化与计算机技术—计算机科学与技术]
  • 作者机构：[1]郑州大学信息工程学院,河南郑州450001, [2]西安电子科技大学计算机学院,陕西西安710071
  • 相关基金：国家自然科学基金资助项目（61070137,60933009）
  • 相关项目：原因学习：理论及在全基因组关联分析中的应用
作者： 耿耀君|张军英|袁细国|
关键词： 极速学习机(ELM), 单隐层前馈神经网络(SLFN), 支持向量机(SVM), 多类分类, 泛化性能, extreme learning machine（ELM）, single hidden layer feedforward neural network（SLFN）, support vector machine（SVM）, multi-classification, generalization performance
中文摘要：

针对极速学习机（ELM）性能过分依赖于隐层节点稠密的单隐层前馈神经网络（SLFN）问题,提出了适用于多类分类的精简型ELM,即SVM-ELM（基于支持向量机优化的ELM）.该方法大幅削减隐层节点数为类别数,同时通过SVM技术优化每个节点的线性决策函数,显著提高单个节点的决策水平,为ELM的宏观决策提供有利条件.在HCL2000,MNIST和USPS等公共数据集上的实验表明：该方法能够减少节点数目而不损害学习精度,当类别数为10时,基于SVM-ELM方法构造的10节点SLFN泛化性能即可超越基于原始ELM方法构造的包含成千上万个隐层节点SLFN的泛化性能.

英文摘要：

A reduced extreme learning machine（ELM）for multi-classification called SVM-ELM（ELM optimized with support vector machine）was proposed for tackling the problem that the single hidden layer feedforward neural networks（SLFN）used by ELM is too large for its applications.The proposed algorithm used only k hidden layer nodes for k-class problem to build the SLFN.Each node was optimized with the SVM technique with a high decision level for the SLFN.Experimental results on HCL2000,MNIST and USPS benchmark datasets show that SVM-ELM could get an excellent generalization performance with fewer nodes.The classification performance of SVM-ELM with 10 hidden layer nodes outperforms that of ELM with thousands of hidden layer nodes.

同期刊论文项目
原因学习理论及在全基因组关联分析中的应用
期刊论文 35 会议论文 4 专利 1
生物网络数据分析与挖掘中相关理论与关键技术
期刊论文 70 会议论文 14 专利 3 著作 1
同项目期刊论文
Performance analysis of novel methods for detecting epistasis
EpiMiner: a three-stage co-information based method for detecting and visualizing epistatic interact
Incorporating heuristic information into ant colony optimization for epistasis detection
Solving Multi-class Learning Problems via Data-driven Topology-Preserving Output Codes
Distributed Dual Cluster Algorithm Based on FCM- Clustering Sensor Streams
Statistical tests based multi-strategy learning method to infer the Gaussian Markov network structur
Quantitative identification of differentially methylated loci based on relative entropy for matched
基于高斯噪声模型的马尔可夫网络构建算法
Robust data clustering by learning multi-metric Lq-norm distances
Probability Theory-Based SNP Association Study Method for Identifying Susceptibility Loci and Geneti
用于FSVM抑噪的统计型模糊权重函数
An overview of population genetic data simulation
一种基于监督降维和形状分析的基因选择方法
Informative energy metric for similarity measure in reproducing kernel hilbert spaces
低剂量CT图像模型参数估计及统计去噪研究
核矩阵列相关低秩近似分解算法
TAGCNA: A method to identify significant consensus events of copy number alterations in cancer
Kernel-kNN:基于信息能度量的核k-最近邻算法
一种基于投影稀疏表示的基因选择方法
基于MCMC方法的自适应低剂量CT图像去噪
长非编码RNA生物特征研究与分析
gene selection for cancer classification using sparse representation coefficient and MRMR criterion
Simulating Linkage Disequilibrium Structures in a Human Population for SNP Association Studies
Comparative Analysis of Methods for Identifying Recurrent Copy Number Alterations in Cancer
An overall-regional competitive self-organizing map neural network for the Euclidean traveling sales
一种鲁棒的子空间聚类算法
基于EM算法的低剂量CT图像去噪
Novel Method of Mining Classification Information for SVM Training
Density-based shrinkage for revealing hierarchical and overlapping community structure in networks
Statistical Sinogram Smoothing for Low-Dose CT with Segmentation-Based Adaptive Filtering
Learning low-rank kernel matrix with column-based methods
从链接密度遍历序列中挖掘网络社团的层次结构
A Computational Method Based on the Integration of Heterogeneous Networks for Predicting Disease-Gen
Seed selection strategy in global network alignment without destroying the entire structures of func
Predicting protein complexes in protein interaction networks using a core-attachment algorithm based
Eigenspaces of networks reveal the overlapping and hierarchical community structure more precisely
一种用于生物网络数据的频繁模式挖掘算法
A degree-distribution based hierarchical agglomerative clustering algorithm for protein complexes id
Identification of overlapping and non-overlapping community structure by fuzzy clustering in complex
ppiPre – an R package for predicting protein-protein interactions
Scaling the Kernel Function based on the Separating Region in Input Space: A Data-dependent Way to I
Biological network analysis: insights into structure and functions
Maximizing modularity intensity for community partition and evolution
Detection of deregulated modules using deregulatory linked path.
Maintain the structural controllability under malicious attacks on directed networks
Weighting links based on edge centrality for community dectection
Semi-supervised clustering algorithm for community structure detection in complex networks
A centrality measure based on spectral optimization of modularity density
Identification of conserved protein complexes by module alignment
Non-traditional spectral clustering algorithms for the detection of community structure in complex n
A network based method for analysis of lncRNA-disease associations and prediction of lncRNAs implica
Identification of core-attachment complexes based on maximal frequent patterns in protein-protein in
Prediction of Human Disease-Related Gene Clusters by Clustering Analysis
Haiyang Wang. Anti-triangle centrality-based community detection in complex networks
核矩阵列相关低秩近似分解算法
Kernel-kNN:基于信息能度量的核k-最近邻算法
一种基于投影稀疏表示的基因选择方法
基于MCMC方法的自适应低剂量CT图像去噪
长非编码RNA生物特征研究与分析
Genome-wide identification of significant aberrations in cancer genome
mirPD: A pattern-based approach for identifying microRNAs from deep sequencing data
Mutual information and linkage disequilibrium based SNP association study by grouping case-control
长非编码RNA特征研究与分析
生物网络比对的模型与算法
Network-Based Inference Framework for Identifying Cancer Genes from Gene Expression Data
An entropy weighting mixture model for subspace clustering of high-dimensional data
非树型网络模体发现算法
基于EM算法的低剂量CT图像去噪
蛋白质网络中复合体和功能模块预测算法研究
动态网络模式挖掘方法及其应用
基于社交圈的在线社交网络朋友推荐算法
一种基于点割的电路划分算法
动态网络桥系数增量聚类算法
复杂网络社团的谱分检测方法
无线传感器网络数据环区域查询处理算法
低剂量CT投影图像噪声分析及去噪算法研究
Novel Method of Mining Classification Information for SVM Training