中文摘要：

情报学领域的研究表明,研究工作者形成的科研合作关系网络是一个非连通网络,但是网络统计参数呈现出无尺度网络特性和小世界网络特性,合作网络是一种描述科研合作的典型网络。提出高阶聚集系数的概念,以期刊作者的合作网络数据为计算案例,计算期刊合作者的高阶聚集系数C（h）,形成图表并分析其高阶聚集系数C（h）在图中所表示的意义。通过采集国内计算机类一级学报的大量合作者样本数据并通过计算分析表明,合作者为20人左右时,合作团队拥有广泛的知识面和较高的合作水平,较易做出高水平的创新性成果,也意味着高水平的论文作者通常和学者20人左右直接或者间接保持合作关系。高阶聚集系数同样可用于语言网络等其他呈现出小世界效应的网络分析。

英文摘要：

The research made in the field of information science shows that scientific collaboration relationship network formed by the researchers is unconnected.While the characteristics of scale-free network and small-world network presented from the network statistical parameter shows that collaboration network is a typical network describing the scientific collaboration research.The concept of high order clustering coefficient is presented.With the use of network data of journal authors,it will calculate journal collaborators＇ high order clustering coefficient C（ h）,and form a chart and analyze the significance of C（ h） in the chart.The analysis of a large amount of collaborator＇sample data collected from the first-class domestic journal of computer shows that when the number of collaborators is around 20,the cooperation team has extensive knowledge and a high level of cooperation and it is easier for the team to make a high level of innovative achievement.That means a high-level author will usually keep a cooperative relationship with about 20 scholars directly or indirectly.