Demonstration of the Importance of System Ingredients with Strong Similarity in Cluster Analysis
In this paper, we propose one method to demonstrate the importance and effectiveness of system ingredients with strong similarity in cluster analysis. As a test, we clustered 1578 SSCI journals with three different collections of journal-journal similarities, which are computed from the aggregated journal-journal citation reports of the Institute of Scientific Informa-tion(ISI). The statistical properties of the clustering results and the consistency of the results with ISI category demonstrate the importance and efficiency of those predominant system ingredients with strong similarly, and may aid the management of information for increasingly large complex systems analysis.
cluster analysis similarity predominant
YunFeng Chang Yuan Zhao ShengQin Feng
College of Science China Three Gorges University YiChang, China
国际会议
北京
英文
309-312
2011-08-08（万方平台首次上网日期，不代表论文的发表时间）