Silhouettes: a graphical aid to the interpretation and validation of cluster analysis
Journal of Computational and Applied Mathematics
Algorithms for clustering data
Algorithms for clustering data
Automatically organizing bookmarks per contents
Proceedings of the fifth international World Wide Web conference on Computer networks and ISDN systems
ACM Computing Surveys (CSUR)
On Clustering Validation Techniques
Journal of Intelligent Information Systems
Cluster Analysis
Journal of the American Society for Information Science and Technology
Survey of clustering algorithms
IEEE Transactions on Neural Networks
Hi-index | 0.00 |
The objective of this paper is to propose a cluster analysis methodology for measuring the performance of research activities in terms of productivity, visibility, quality, prestige and international collaboration. The proposed methodology is based on bibliometric techniques and permits a robust multi-dimensional cluster analysis at different levels. The main goal is to form different clusters, maximizing within-cluster homogeneity and between-cluster heterogeneity. The cluster analysis methodology has been applied to the Spanish public universities and their academic staff in the computer science area. Results show that Spanish public universities fall into four different clusters, whereas academic staff belong into six different clusters. Each cluster is interpreted as providing a characterization of research activity by universities and academic staff, identifying both their strengths and weaknesses. The resulting clusters could have potential implications on research policy, proposing collaborations and alliances among universities, supporting institutions in the processes of strategic planning, and verifying the effectiveness of research policies, among others.