Constrained K-means Clustering with Background Knowledge
ICML '01 Proceedings of the Eighteenth International Conference on Machine Learning
Laplacian Eigenmaps for dimensionality reduction and data representation
Neural Computation
Enhancing semi-supervised clustering: a feature projection perspective
Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining
A tutorial on spectral clustering
Statistics and Computing
Pairwise constraint propagation by semidefinite programming for semi-supervised classification
Proceedings of the 25th international conference on Machine learning
A graph-based projection approach for semi-supervised clustering
PKAW'10 Proceedings of the 11th international conference on Knowledge management and acquisition for smart systems and services
Influence of erroneous pairwise constraints in semi-supervised clustering
AMT'12 Proceedings of the 8th international conference on Active Media Technology
| Hi-index | 0.00 |
This paper proposes an approach for pairwise constraint propagation in graph-based semi-supervised clustering. In our approach, the entire dataset is represented as an edge-weighted graph by mapping each data instance as a vertex and connecting the instances by edges with their similarities. Based on the pairwise constraints, the graph is then modified by contraction in graph theory to reflect must-link constraints, and graph Laplacian in spectral graph theory to reflect cannot-link constraints. However, the latter was not effectively utilized in previous approaches. We propose to propagate pairwise constraints to other pairs of instances over the graph by defining a novel label matrix and utilizing it as a regularization term. The proposed approach is evaluated over several real world datasets, and compared with previous regularized spectral clustering and other methods. The results are encouraging and show that it is worthwhile to pursue the proposed approach.