基于激活力的复杂网络建模及其应用

项目名称： 基于激活力的复杂网络建模及其应用

项目编号： No.61273217

项目类型： 面上项目

立项/批准年度： 2013

项目学科： 自动化技术、计算机技术

项目作者： 郭军

作者单位： 北京邮电大学

项目金额： 82万元

中文摘要： 申请人通过对多类复杂网络节点关系的分析，新近提出了一种可普遍应用的表达网络节点关系的统计量-激活力和与其紧耦合的节点亲和度测度，获得了一种对复杂网络建模的新方法，并成功应用于词网和蛋白质相互作用网络的建模分析。本项目在上述工作的基础上研究复杂网络的基本建模方法及其在图像和语音建模以及癌症基因组分析等重要问题中的应用。主要创新点包括:1)创立和完善基于激活力对复杂网络进行建模的基础理论，为各类复杂网络的分析提供有效的新方法;2)用所创立的复杂网络模型对图像和语音进行建模，对图像和语音中特征的结构关系进行描述，为分类、聚类、标注、检索等应用提供更有效的支撑;3)用所创立的复杂网络模型对癌症基因组进行分析，以期揭示导致癌症的突变基因之间的相互诱导关系，获得对癌症形成机制更深的理解。

中文关键词： 激活力；复杂网络模型；局部共现建模；癌症基因组分析；结构化稀疏建模

英文摘要： Through analyzing the relations between the nodes in various complex networks, the applicant recently presented a new sort of statistics-activation forces, which can be widely applied to describe the relations between nodes, and a measure of affinity between nodes that tightly couples with the activation forces. This provides a new approach to modelling complex networks and has been successfully applied in the analyses of word networks and protein-protein interaction networks. Based on the above work, this project is going to study the fundamentals of complex netwrok modelling and the applications of image and voice modelling and cancer genome analysis. Major research targets include: 1) to create and perfect the basic theories of modelling complex netwroks based on the activation forces, providing an even more effective approach to analyzing various types of complex networks; 2) by using the created model of complex networks to represent images and voices, describing the structural relations between features in images and voices to provide better support to the applications of classification, clustering, labelling and retrieval; 3) by using the created model of complex networks to analyze cancer genome, in order to reveal the relations between gene mutations which cause cancers and to acqure more understanding

英文关键词： Activation Forces；Modeling Complex Networks；Modelling Local Co-occurrence；Cancer Genome Analysis；Structured Sparse Modeling