基于社会网络分析方法的中国计算机领域科研论文合作规律探析

发布时间：2019-01-27 06:30

【摘要】：如果将论文作者看成网络上的一个结点，当他们共同发表文章时，他们之间有一条边相连，这样就构成了论文合作网络。本质上讲，论文合作网是一种特殊的社会网络。著名学者M.E.J.Newman认为，论文合作网络是指在进行科学研究过程中，，科研工作者由于科研工作的需要与其它科研人员进行交流、合作所形成的一种关系网络。我们以2000年至2011年期间我国计算机领域20种核心期刊为数据源，构造了计算机领域2000至2011期间的论文合作网络；构造了每种期刊2000至2011期间的论文合作网络；分类构造了计算机科学和软件技术类期刊论文合作网络、计算机图像图形学类期刊论文合作网络、计算机自动化与测控类期刊论文合作网络、传感器理论和应用类期刊论文合作网络。采用社会网络分析方法从网络个性属性、整体属性的角度分析了上述网络的特点，并得出以下结论和规律：（1）论文合作网络是无标度网络，具有比较明显的的小世界效应，核心结点具有高度值（Large Degree）和高中介性（Large Betweenness）的特点。（2）论文合作网络具有多核网络结构。网络均不是一个全连通的网络，每一个网络都存在一些小团体，这些小团体多数是国家级重点实验室或研究所。（3）论文合作网具有明显的内部核心网和外围网。核心元素大多数来自于内部核心网；有些外围网元素的中介性指标和度值比较大，说明这些外围网元素在网络连接上起到了一定的作用。（4）核心元素均为度和中介性比较大的科研人员；同一个网络中，度和中介性排列顺序并不完全相同；中介性排名靠前的结点其度不一定靠前，这些人员大都是跨学科、跨领域的科研人员。（5）计算机科学和软件技术类论文合作网的平均路径较大，平均聚类系数较小，具有一定的小世界效应，但不太明显。（6）网络存在正常的离散结点，这些离散结点多数为论文合作网络的新人。最后，本文还提出了建设一个高效的小团体科研网的一些建议和做法。
[Abstract]:If the authors are regarded as a node in the network, when they publish a paper together, they are connected with each other by an edge, thus forming a cooperative network of papers. In essence, the paper cooperation network is a special social network. M.E.J.Newman, a famous scholar, believes that the cooperative network of papers is a kind of relationship network formed by the scientific research workers to communicate with other researchers and cooperate in the process of carrying out scientific research. From 2000 to 2011, we take 20 kinds of core journals in computer field as data source, construct the paper cooperation network between 2000 and 2011 in computer field, construct the paper cooperation network between 2000 and 2011 in each journal. The thesis cooperation network of computer science and software technology, computer image graphics, computer automation and measurement and control are classified. Cooperative Network of Sensor Theory and Application Journals. The characteristics of the above mentioned networks are analyzed by using the social network analysis method from the angle of the network personality attribute and the whole attribute, and the following conclusions and rules are obtained: (1) the cooperative network in this paper is scale-free network. The core nodes have the characteristics of high value (Large Degree) and high intermediate (Large Betweenness). (2) the cooperative network has multi-core network structure. The network is not a fully connected network. There are some small groups in each network, most of which are national key laboratories or research institutes. (3) the thesis cooperation network has obvious inner core network and peripheral network. Most of the core elements come from the inner core network, and some peripheral network elements play a certain role in the network connection. (4) in the same network, the order of degree and intermediation is not exactly the same; The intermediate rank is not necessarily at the top. Most of these people are interdisciplinary and interdisciplinary researchers. (5) the average path of the cooperative network of computer science and software technology is large, the average clustering coefficient is small, and it has a small world effect, but it is not obvious. (6) there are normal discrete nodes in the network, most of which are new to the cooperative network. Finally, some suggestions and practices for constructing an efficient small group scientific research network are put forward.
【学位授予单位】：江西师范大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：O157.5

【参考文献】