Title
Academic co-author networks based on the self-organizing feature map
Abstract
This paper studies mathematician Paul Erdös, one of the most famous academic co-authors and his large co-author network. K-means and Self-Organizing feature Map (SOM) algorithms are applied to study the network. First, the SOM algorithm is introduced to recognize the pattern area number, which can identify the sub-segment automatically and export the central point of each cluster as well as the weights. Taking the results as the initial input of the K-means algorithm to make the further clustering, the accurate clustering results are gained. Then search the largest weight node of each cluster and set it as the most influential researcher. Finally we compared the h-index of the most influential researcher with the corresponding weight node of the cluster, the results confirm that the algorithm is better than SOM and K-means algorithms when they are separately used.
Year
DOI
Venue
2014
10.1109/FSKD.2014.6980858
FSKD
Keywords
Field
DocType
co-author networsk,pattern area number,weight,k-means,pattern clustering,som,som algorithms,h-index,academic coauthor networks,k-means algorithms,influential researcher,self-organising feature maps,self-organizing feature map algorithms
Pattern recognition,Computer science,Self,Artificial intelligence,Machine learning
Conference
Citations 
PageRank 
References 
0
0.34
6
Authors
7
Name
Order
Citations
PageRank
Gege Zhang100.34
Wei-Xing Zhou220615.05
Yuanyuan Zhang312111.56
Xiao-Hui Hu4105.55
Yun Xue565.30
jianping wang6405.87
Meihang Li783.83