摘要: In this paper, we study the ill-effects of bridgenodes, which causes many dissimilar objects to be placed together in same cluster by existing clustering algorithms. We offer two new metrics for measuring how well a algorithm handles presence bridge-nodes. also illustrate algorithms that produce overlapping clusters help alleviate effect bridge-nodes and form more meaningful clusters. However, if there is too much overlap, become less informative. To address problem, present novel called MIN-CUT. Our experimental results with real data sets show MIN-CUT leads purer have very little overlap.
参考文章(14)
S. S. Ravi, Ian Davidson, Towards Efficient and Improved Hierarchical Clustering With Instance and Cluster Level Constraints ,(2005)
Michael Randolph Garey, Johnson: "computers and intractability ,(1979)
Michael Randolph Garey, David S. Johnson, A guide to the theory of np-completeness ,(1978)
Vipin Kumar, Pang-Ning Tan, Michael M. Steinbach, Introduction to Data Mining ,(2013)
Gary Ortrud R Chartrand, Applied and algorithmic graph theory ,(1992)
S. S. Ravi, Ian Davidson, Clustering with Constraints: Feasibility Issues and the k-Means Algorithm. siam international conference on data mining. pp. 138- 149 ,(2005)
Richard C. Dubes, Anil K. Jain, Algorithms for clustering data ,(1988)
Michael R. Garey, David S. Johnson, Computers and Intractability: A Guide to the Theory of NP-Completeness ,(1979)
Mikhail Bilenko, Sugato Basu, Raymond J. Mooney, Integrating constraints and metric learning in semi-supervised clustering international conference on machine learning. pp. 11- ,(2004) , 10.1145/1015330.1015360
Roger N. Shepard, Phipps Arabie, Additive clustering: Representation of similarities as combinations of discrete overlapping properties. Psychological Review. ,vol. 86, pp. 87- 123 ,(1979) , 10.1037/0033-295X.86.2.87