Clustering Algorithm For Relational Data Research Articles

Q-Divergence-Based Relational Fuzzy c-Means Clustering

In this paper, a clustering algorithm for relational data based on q-divergence between memberships and variables that control cluster sizes is proposed. A conventional method for vectorial data is first presented for interpretation as the regularization of another conventional method with q-divergence. With this interpretation, a clustering algorithm for relational data, based on q-divergence, is then derived from an optimization problem built by regularizing the conventional method with q-divergence. A theoretical discussion reveals the property of the proposed method. Numerical results are presented that substantiate this property and show that the proposed method outperforms two conventional methods in terms of accuracy.

Nonlinear multicriteria clustering based on multiple dissimilarity matrices

We present a new algorithm capable of partitioning sets of objects by taking simultaneously into account their relational descriptions given by multiple dissimilarity matrices. The novelty of the algorithm is that it is based on a nonlinear aggregation criterion, weighted Tchebycheff distances, more appropriate than linear combinations (such as weighted averages) for the construction of compromise solutions. We obtain a hard partition of the set of objects, the prototype of each cluster and a weight vector that indicates the relevance of each matrix in each cluster. Since this is a clustering algorithm for relational data, it is compatible with any distance function used to measure the dissimilarity between objects. Results obtained in experiments with data sets (synthetic and real) show the usefulness of the proposed algorithm.

Weighted fuzzy clustering algorithm for relational data with-multiple medoids

It is apparently inadequate that k center algorithm uses only one point to represent the whole class,which will definitely affect the accuracy of clustering results.Therefore,a weighted fuzzy clustering algorithm for relational data with multiple medoids was proposed.In the proposed algorithm,multiple objects in each cluster carried different weights called medoids weights to represent their degrees of representativeness in that cluster.This algorithm can make each cluster to be represented by multiple objects instead of only one object.The experimental results show that the proposed algorithm can capture the underlying structures of the data more accurately and provide richer information for the description of the resulting clusters.