生命科学近五年论文引文情况分析  

Analyzing the Structure and Evolution of Recent Five Years Bibliographies in Life Science

作　　者：杨胜琦[1] 吴斌[1] 

机构地区：[1]北京邮电大学智能通信软件与多媒体北京市重点实验室,北京100876

出　　处：《数字图书馆论坛》2008年第6期12-20,共9页Digital Library Forum

基　　金：本文得到国家自然科学基金项目（60402011）和国家科技支撑计划项目（2006BAH03805）资助.致谢：感谢韩超、胡德勇、操牡丹、周丹和龙海燕,他们在本文前期的数据处理过程中,给予我们极大的帮助.感谢叶祺,在形成本文的过程中,我们之间有过多次愉快而有意义的讨论.

摘　　要：基于网络描述的复杂社会结构能够更好地展示网络中个体的联系特征，由此产生的复杂网络理论已经被广泛应用到社会科学的各个领域。近年来，除了对网络结构所具有的小世界、幂率分布等静态特性的分析外，大量研究开始关注网络结构中个体的组织特征。由这些个体组成的子图中，个体间有着更高的连接特征，而与其他子单元间的个体连接则相对稀疏。这种子单元通常被称为社团。社团发现及分析对研究网络的组织结构和社会特征有着重要意义。将社团发现方法应用到文献分析中，可以得到各学科领域的特征及关联关系。文章利用生命科学领域最近五年间的期-{iJ论文文摘记录，构造了两种引文网络。直接的引用网络和间接的论文耦合网络。对这两个网络基本属性的分析有助于了解生命科学领域发展的现状。此外，文章还使用了两种基于耦合网络的社团分析方法，重点分析了最近五年间生命科学领域的学科分类、关联特征以及随时间的演化情况，以助于理解整个生命科学领域的学科结构。Network. based complex social structure can well capture the intricate connection properties of every individual. The resulted complex network. theory is now widely used in areas of social science. Recently, besides the static statistic in the small-world property and power-law degree distributions, researches have concentrated particularly on the global organization of individuals in such networks. In some sub-modules composed from specific individuals, there are more inter-connectione, opposite to the looser outer-connections. Such submodule is usually called community （or cluster, group）. The detection and analysis of community structure in network, is a topic of considerable recent interest and is crucial to interpret the global organization and social characteristics of such networks. Applying this method to bibliographies, we can obtain the characteristics and the associations of subjects in specific area. In this article, using the data from latest five years bibliographies in life science, we construct two kinds of network.. One is a directed citing network., the other is a couping one. According to the statistic analysis of these two network.s, we can get the basic knewledge of the recent development in life science. Moreover, we put highlights on the analysis of community. During which section, we construct two analytical community methods based on the coupling graph. The simple hierarchical aggregation provides a survey of communities at a global level, the CNM algorithm in coupling network, effectively reveals its subjects classification, association and evolution.

关 键 词：复杂网络 生命科学 社团 

分 类 号：C95[社会学—民族学] Q1-0[生物学—普通生物学]