cluster property meaning in Chinese
结团性质
Examples
- Based on unsupervised learning , sparse coding is suitable to describe images with non - gaussian distribution and can get rid of the high order redundancy among the image pixels . since the basis function of sparse coding has build - in clustering property , it increases the inter - class variations of the features
稀疏编码是一种基于非监督学习的算法，它适合描述具有非高斯分布的数据对象，能够有效地消除图像象素点之间的冗余，并具有内在的聚类特性。 - Based on this kind of relations between the topological structures and the content distributions we study the web modelling , community identification and some related application problems in detail : first , after some existed characteristics of the web topology are verified , some new characteristics are discovered : the high clustering property in micro - topology ( high average gathering coefficient ) , the obvious mapping relation between the topological struture and the content in micro - level 、 linear irrelevant between the degree distribution of network nodes and the relative degree distribution of contents etc . then after analysis the topology of the complex network and the network modeling , the muti - scale determinism is proposed , especially for the information network a web evolvement model ( prcp model ) that fused the node authority and the node correlation is proposed . the model deduction , evolving learning verification and large scale experiment proof indicate that the model can explain the micro - topology centralizing phenomena , can imitate the mapping relation between the network connecting distribution and network content relative distribution and also can predict the mapping relation between the topology clustering and content clustering
本文在详细观察了web网络的拓扑结构特征以及拓扑结构与内容分布相互关系的基础上,以信息网络的物理连接拓扑结构与节点内容相关度分布之间的相互关系为主线,从网络特征、网络建模、社区分析及相关应用方面问题进行了深入细致地探讨:首先在验证了前人提出的web网络拓扑结构特征基础上,进一步发现了信息网络所具有的一些新特征: 1 )网络微观颗粒度的拓扑结构聚团与内容聚团存在明显的映射关系,具体包括节点之间的物理连边概率与节点之间的内容相关度成指数比例关系、节点形成三角形拓扑结构的概率与节点内容相关紧密程度之间同样具有一种指数比例关系; 2 )网络节点连接度整体分布与节点内容相关度整体分布是线性无关的; 3 )网络微观拓扑结构中的存在很强的集聚性(平均聚团系数很高) 。 - Based on the clustering property of the basis function of sparse coding , a basis function initialization method using fuzzy c mean algorithm is proposed to help the energy function of sparse coding to converge to a better local minimum for recognition . experimental results show that the classification and the sparseness of the features are both improved
经过模糊c均值聚类初始化后的基函数能够让稀疏编码的能量函数收敛到一个更有利于识别的局部最小点，试验结果表明特征的分类性和稀疏性都得到了提高。